- Email: [email protected]
Safer colonoscopy with patient-controlled analgesia and sedation with propofol and alfentanil
Safer colonoscopy with patient-controlled analgesia and sedation with propofol and alfentanil Daniel Külling, MD, Amedeo C. Fantin, MD, Peter Biro, MD, Peter Bauerfeind, MD, Michael Fried, MD Zürich, Switzerland
Background: The aim of this study was to assess the efficacy of patient-controlled analgesia and sedation with propofol/alfentanil for colonoscopy compared with continuous drug infusion and conventional nurse-administered medication. Methods: One hundred fifty patients undergoing colonoscopy on an outpatient basis were randomly assigned to 1 of 3 medication regimens. To maintain blinding, all patients were connected to an infusion pump. Group I patients could self-administer boluses of 4.8 mg propofol and 125 µg alfentanil without restriction. Group II patients received a continuous infusion with 0.048 mg/kg propofol and 0.12 µg/kg alfentanil per minute. Group III patients received intravenous premedication with 0.035 mg/kg midazolam and 0.35 mg/kg meperidine. Results: There were no differences between the groups with respect to pain (visual analogue scale) and procedure time. Patient-controlled analgesia and sedation with propofol/alfentanil (group I) resulted in less of an increase in the transcutaneous partial pressure of carbon dioxide (p = 0.0004) during colonoscopy and less of a decrease in mean arterial blood pressure (p = 0.0021) during recovery, as well as more complete recovery (p = 0.0019) after 45 minutes compared with conventional administration of midazolam/meperidine. Furthermore, patient-controlled analgesia and sedation yielded a higher degree of patient satisfaction than continuous infusion of propofol/alfentanil (p = 0.0033) or nurse-administered midazolam/meperidine (p = 0.0094). Conclusions: Patient-controlled administration of propofol and alfentanil for colonoscopy may provide a better margin of safety than conventional administration of midazolam and meperidine and results in a higher level of patient satisfaction and shorter recovery. (Gastrointest Endosc 2001;54:1-7.)
Various drugs are used to achieve sedation and analgesia during colonoscopy. Most endoscopists use sedative hypnotics such as midazolam or diazepam, alone or in combination with opiate narcotics, usually meperidine.1 Although some patients tolerate colonoscopy without sedation, most experience some discomfort, which may influence the successful completion of the examination. Variations in pain threshold, drug tolerance, and colonic sensitivity among individual patients make it difficult to anticipate the required dose of sedative medication. As a consequence, some patients are given an inadequate dose, whereas others are oversedated with the associated risk of cardiorespiratory complications and delayed recovery.2 Furthermore, patient versus physician assessment of discomfort during proceReceived November 28, 2000. For revision January 31, 2001. Accepted April 3, 2001. From the Division of Gastroenterology, Department of Internal Medicine, and the Institute of Anesthesiology, University Hospital of Zürich, Switzerland. Reprint requests: Daniel Külling, MD, Division of Gastroenterology, Department of Internal Medicine, University Hospital, Rämistr. 100, CH-8091 Zürich, Switzerland. Copyright © 2001 by the American Society for Gastrointestinal Endoscopy 0016-5107/2001/$35.00 + 0 37/1/116174 doi:10.1067/mge.2001.116174 VOLUME 54, NO. 1, 2001
dures differs significantly.3,4 To overcome the problem of individual differences in the perception of pain, patient-controlled analgesia (PCA) delivery systems were introduced as early as 1968. Since then, the results of many trials support the safety and efficacy of PCA for various pain syndromes including those relating to surgery, cancer, trauma, and obstetric procedures.5 Reviews of PCA indicate that patient acceptance is generally enthusiastic because of the possibility of self-control.6,7 The hypnotic and sedative agent propofol (2,6diisopropylphenol) and the opiate alfentanil both have rapid onset and termination of action, making them ideal choices for patient-controlled administration. Patient-controlled analgesia and sedation (PCAS) with propofol alone or in combination with alfentanil has been used during hip and knee arthroplasty, oral surgery, transvaginal oocyte retrieval, and also colonoscopy.8-14 These studies concluded that PCAS with propofol/alfentanil provides safe and satisfactory sedation and shortens recovery time. The present randomized doubleblind controlled trial compares PCAS with propofol and alfentanil with nurse-administered midazolam and meperidine and with continuous infusion of propofol and alfentanil without patient control in patients undergoing colonoscopy. GASTROINTESTINAL ENDOSCOPY
1
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
Table 1. Study design Group
I
Initial bolus Pump content Pump setting Additional boluses
Placebo (saline solution) Propofol + alfentanil PCAS Propofol + alfentanil
MATERIALS AND METHODS Patients and study design One hundred fifty consecutive patients scheduled for colonoscopy on an outpatient basis were enrolled in the study. Indications for colonoscopy included abdominal pain (39%), rectal bleeding or anemia (23%), altered bowel habits (17%), follow-up after polypectomy (16%), and miscellaneous (5%). Exclusion criteria were age below 18 or above 75 years, a history of colonic resection, allergy to study medication, severe cardiorespiratory disease (ASA classification III or higher),15 renal or hepatic impairment, seizure disorder, psychiatric disease, substance abuse, pregnancy or breast-feeding, and refusal to receive any medication for the forthcoming colonoscopy. All colonoscopies were performed by expert endoscopists who possessed equal levels of skill. According to a randomization list, patients were assigned to PCAS with propofol and alfentanil (group I), continuous infusion of propofol and alfentanil (group II), or conventional sedation with midazolam and meperidine (group III) (Table 1). Patients as well as endoscopists were blinded as to the sedation regimen being used. However, endoscopists may have been aware of patient attempts to depress the PCAS button to control discomfort, which could have biased endoscopist requests for additional boluses of medication. Patient information, preparation of the study medication, and patient monitoring during and after colonoscopy were performed by a gastroenterologist and/or a nurse who were not involved in the procedure. The study personnel were trained in cardiopulmonary resuscitation. An anesthetist was not present in the procedure room, but anesthesia support was on call at all times. The study was approved by the ethics committee of our hospital. Written informed consent was obtained. During this process, the education level of the patient was recorded and patients were asked to rate their anxiety in relation to the expected colonoscopy on a 10-cm visual analogue scale (VAS) (left end point “no anxiety,” right end point “very anxious”). Medication To maintain blinding, all patients received an intravenous bolus injection at the beginning of the colonoscopy (saline solution in group I and II, midazolam and meperidine in group III) (Table 1). Furthermore, all patients were connected to a PCA pump. They were instructed to depress the button as often as required to relieve discomfort. For patients in group I and II, the pump contained propofol and alfentanil; for group III it contained saline solution. In group I, the pump was programmed in PCAS mode, in group II and III for constant continuous infusion. 2
GASTROINTESTINAL ENDOSCOPY
II
III
Placebo (saline solution) Propofol + alfentanil Continuous infusion Propofol + alfentanil
Midazolam + meperidine Placebo (saline solution) Continuous infusion Midazolam or meperidine
As regulated by the manufacturers, propofol (10 mg/mL, Ansiven, Abbott, Baar, Switzerland) was mixed with alfentanil (0.5 mg/mL, Rapifen, Janssen-Cilag, Baar, Switzerland) at a volume-ratio of 20:1.16 In group I, the PCA pump (Graseby 3300, Graseby Medical, Watford, UK) was programmed to deliver a bolus dose of 0.5 mL (4.8 mg propofol and 12 µg alfentanil) and with a zero lockout interval.9,11,13,14 In group II, the pump was programmed at a constant continuous infusion rate of 0.005 mL/min × body weight in kg.11,15 In groups I and II, the study-nurse could administer additional 1-mL boluses whenever the endoscopist felt that the patient was uncomfortable and required additional sedation up to a maximal total dosage of 30 mL (288 mg propofol and 0.72 mg alfentanil). In group III, midazolam (1 mg/mL, Dormicum, Roche Pharma, Reinach, Switzerland) 0.035 mg/kg body weight and meperidine (50 mg/mL, Pethidin, Streuli, Uznach, Switzerland) 0.35 mg/kg body weight were administered intravenously before the start of the colonoscopy.1,2 During the procedure, alternating boluses of midazolam 0.015 mg/kg or meperidine 0.35 mg/kg were given as needed up to a maximal total dosage of 0.08 mg/kg midazolam and 0.7 mg/kg meperidine. Assessment The primary study objective was patient satisfaction with the degree of sedation during colonoscopy. Secondary parameters included pain, procedure time, amnesia, safety, and recovery. Ninety minutes after the end of the colonoscopy, the patients assessed their pain during the colonoscopy on a 10-cm VAS (left end point “no pain,” right end point “severe pain”) and their satisfaction with the degree of sedation (left end point “ideal sedation,” right end point “completely inadequate sedation”). Still without knowledge of the drug regimen they had received, the patients were also questioned as to whether they would request the same method of analgesia and sedation for a future colonoscopy (yes or no). In addition, the patients were given a questionnaire to be filled out the next day that asked about after-effects severe enough to disrupt their regular activities. Amnesia was tested by questioning the patients 90 minutes after the end of the procedure about the word (“peach”) they had been told to memorize during the colonoscopy and before withdrawal of the colonoscope. Patients were monitored during colonoscopy and recovery by clinical observation, continuous pulse oximetry, and automated sphygmomanometry at 5- and 15-minute intervals, respectively. All patients were given oxygen 2 L/min by nasal cannula during the entire procedure. During recovery, nasal oxygen supplementation was given VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
D Külling, A Fantin, P Biro, et al.
Table 2. Patient characteristics Group
I
Age (y) Male/female College degree Previous colonoscopies with/without sedation Anxiety before colonoscopy (10 cm VAS score)
II
55 (43-63) 56%/44% 22% 22%/18% 2.0 (0.0-5.0)
55 (40-65) 44%/56% 16% 30%/24% 3.0 (1.0-5.0)
III 48 (35-64) 54%/46% 24% 20%/24% 3.0 (0.0-5.0)
Data are expressed as medians (interquartile ranges) or percentages. There are no significant differences among the 3 groups.
for desaturation to less than 90%. Furthermore, patients were continuously monitored during colonoscopy by a transcutaneous carbon dioxide tension sensor (M1205A Viridia, Hewlett Packard, Andover, Mass.).17 Recovery from sedation was judged by using the Trieger dot-joining test before (baseline score), as well as 15, 45, and 90 minutes after colonoscopy. The Trieger test, which was adapted from the Bender Motor Gestalt test, measures sensory motor performance, a critical determinant of recovery. This test was first prospectively evaluated in 1969 and proved to be a simple reliable method to assess recovery after intravenous administration of sedative drugs or general anesthesia.18 The Trieger test has since been applied in several trials, including 2 recent endoscopic studies.12,19 Forty-two dots at 10- to 15-mm intervals are placed in the form of a 3-sided figure and a transversely placed bellshaped curve, with the 2 figures sharing 1 dot. The patients were asked to connect the dots with a pen. Scoring was calculated by multiplying the number of missed dots with the total error magnitude in millimeters. Statistical analysis and power calculation The medians and interquartile ranges (IQR) were calculated for quantitative variables. Mann-Whitney U rank sum tests were applied to compare the results from the 3 patient groups. Qualitative data were compared with Chisquare tests including Yates’ correction for continuity. A p value of less than 0.05 was considered significant. Power calculations were based on the primary objective of the study, which was patient satisfaction with the degree and quality of sedation. It was determined that 50 patients were needed in each group to detect a difference of 2 units or greater with a level of significance of 0.05, a power of 80%, and an expected variation of 3 units.
RESULTS Patients There were 77 men and 73 women, median age of 54 years (IQR 39-63), enrolled in the study. Another 16 patients who fulfilled entry criteria declined to participate. The 3 patient groups were similar with respect to age, gender, education level, previous colonoscopies with or without sedation, and precolonoscopy anxiety (Table 2). Medication Patients in the PCAS group (I) depressed the handheld button a median of 24 (IQR 16-45) times VOLUME 54, NO. 1, 2001
with 69% (IQR 53%-89%) being successful attempts that actually delivered 0.5-mL boluses. There was no difference with regard to the total number of attempts between groups II (23, IQR 12-51) and III (17, IQR 7-36). Both of the propofol/alfentanil groups received similar total drug doses. The patients in PCAS group I self-administered a median of 8.25 mL (IQR 6.0-12.5) or 78 mg (IQR 57-119) propofol and 198 µg (IQR 144-300) alfentanil, whereas group II received 9.45 mL (IQR 7.2-13.8) (equivalent to propofol 90 mg, IQR 68-131 and alfentanil 227 µg, IQR 173-331) by continuous infusion. This amount included one to four 1-mL nurseadministered boluses in 36% of group II patients, whereas no nurse-administered boluses were given to any group I patient. The median dosages of midazolam and meperidine in group III were 2.7 mg (IQR 2.3-3.0) and 27 mg (IQR 23-30), respectively. This amount included 1 or 2 additional boluses of medication during the procedure in 20% of group III patients. Satisfaction and amnesia Median procedure duration ranged from 22 to 25 minutes without significant differences between the 3 groups. Total colonoscopy with ileal or cecal intubation was achieved in all but 1 patient of group I, this procedure being abandoned in the sigmoid colon because of stenosing diverticulosis. All patients reported similar median pain scores (group I: 4.0, IQR 2.0-5.0; group II: 4.5, IQR 2.5-6.0; group III: 3.0, IQR 1.0-5.0) (Fig. 1). A higher percentage of patients medicated with midazolam/meperidine (50%) demonstrated amnesia for the period of the colonoscopy compared with patients who received propofol/alfentanil (24% in group I, p = 0.0129; 30% in group II, p > 0.05). Still, patients in the PCAS group (I) were more satisfied with the degree of sedation (10, IQR 9.0-10.0) than those in group II (8.5, IQR 6.0-10.0, p = 0.0033) and group III (8.5, IQR 6.510.0, p = 0.0094). Two patients in group III were not satisfied because the degree of sedation was too profound. Otherwise, all dissatisfied patients complained of inadequate sedation. At subgroup analysis, gender, education level, number of prior colonoscopies, indication for colonoscopy, and preGASTROINTESTINAL ENDOSCOPY
3
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
Figure 1. Visual analogue scores for pain during colonoscopy and patient satisfaction with the degree of sedation. Data are expressed as medians and interquartile ranges. Asterisk indicates p = 0.0033 for comparison of group I with group II; p = 0.0094 for comparison of group I with group III.
colonoscopy anxiety did not significantly influence the satisfaction advantage of PCAS. Still blinded as to drug regimen, more patients in group I (94%) than in group II (78%, p = 0.0437) and III (74%, p = 0.0141) indicated they would request the same method of sedation again in the event that a colonoscopy would be required in the future. Safety There were no complications related to the colonoscopy procedures. No serious respiratory or hemodynamic complications occurred with any drug regimen. During colonoscopy, there was no difference among the 3 groups with respect to changes in mean arterial blood pressure (Fig. 2) or oxygen saturation. Two patients in group I, 1 in group II, and none in group III had oxygen saturations between 88% and 90% with prompt normalization after increasing nasal oxygen supplementation from 2 to 4 L/min. PCAS patients exhibited less of an increase in transcutaneous partial pressure of carbon dioxide (pCO2) values (2.6 mm Hg, IQR 0.7-4.5) compared with groups II (4.1 mm Hg, IQR 3.0-8.3, p = 0.0120) and III (5.3 mm Hg, IQR 3.0-8.3, p = 0.0004). The changes in pulse rate and oxygen saturation during recovery were similar for all 3 groups. One patient in group I, 2 in group II, and 1 in group III required nasal oxygen delivery because of desatura4
GASTROINTESTINAL ENDOSCOPY
Figure 2. Safety. Increase in transcutaneous pCO2 during colonoscopy and decrease in mean arterial blood pressure (MABP) during colonoscopy and recovery. Data are expressed in mm Hg as medians and interquartile ranges. Asterisk indicates p = 0.0120 for comparison of group I with group II; p = 0.0004 for comparison of group I with group II. Number sign indicates p = 0.0021 for comparison of group I with group III.
tion to less than 90% with the lowest level being 86%. The patients who received midazolam/meperidine had a more accentuated decrease in mean blood pressure (8.5 mm Hg, IQR –1.0-15.0) than patients in the PCAS group (3.5 mm Hg, IQR –3.0-7.0, p = 0.0021). Recovery Fifteen minutes after the procedure, there was no difference between the 3 groups with regard to recovery from sedation as assessed by the Trieger dot-joining test (Fig. 3). However, at 45 and at 90 minutes, respectively, the Trieger scores of patients in the PCAS group were closer to baseline (0, IQR –0.5-0.5 and 0, IQR –1.0-0.5) than the scores of the patients who received midazolam/meperidine (1.0, IQR 0.0-9.0, p = 0.0019 and 0.25, IQR 0.0-1.5, p = 0.0469). Furthermore, more group III patients (36%) complained of after-effects during the 24 hours after the procedure than group I patients (11%, p = 0.006). The recovery scores of group II patients were VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
D Külling, A Fantin, P Biro, et al.
in between the results for group I and III patients without reaching statistical significance with respect to either group. DISCUSSION The results of this trial indicate that patientcontrolled analgesia and sedation with propofol and alfentanil during colonoscopy may result in better cardiorespiratory function and more rapid recovery compared with conventional nurse-administered midazolam and meperidine. Furthermore, PCAS with propofol/alfentanil yielded a higher degree of patient satisfaction than continuous infusion of the same drugs or nurse-administered midazolam/meperidine. The results of 3 randomized controlled endoscopic trials of PCAS are available, 2 with colonoscopy4,14 and 1 ERCP.3 In the studies by Jowell et al.3 and Stermer et al.,4 baseline sedation was induced with midazolam and meperidine in the “PCAS” study group with drug dosages that exceed the total doses given to our group III patients. Furthermore, the lockout interval (minimum time between doses) of the PCAS pump containing midazolam and meperidine was set to 5 minutes. Under these circumstances, it is unrealistic to assume a patient-controlled setting for a procedure such as colonoscopy in which the painful insertion phase usually lasts only 5 to 20 minutes. Therefore, it is not surprising that these 2 studies did not demonstrate any benefit of PCAS compared with standard analgesia and sedation.3,4 Similar to the present study, Roseveare et al.14 demonstrated that sole patient-controlled application of propofol and alfentanil during colonoscopy provided greater safety than conventional premedication with meperidine and diazepam. Propofol is at least as effective as midazolam for sedation in GI endoscopy and has a considerably faster onset and termination of action.19-22 A pilot study demonstrated that patient-controlled administration of propofol alone did not provide adequate pain relief for colonoscopy, but the addition of the opiate, alfentanil, offered adequate pain control and patient tolerance for the procedure.12 Based on published data,13,14 the lockout time of the PCAS pump was set at zero in the present study. Because the pump delivered the drug at a rate of approximately 200 mL/h, a 0.5-mL bolus (containing 4.8 mg of propofol and 12 µg alfentanil) was administered every 9 seconds if the hand-held button was repeatedly pressed. For comparison, single-induction doses for general anesthesia are 140 mg of propofol and up to 8.5 mg of alfentanil for a 70-kg patient.16 Even with the fast delivery system and rapid action of the drugs, there is necessarily a short delay after VOLUME 54, NO. 1, 2001
Figure 3. Recovery. Trieger scores above baseline 15, 45, and 90 minutes after colonoscopy. Data are expressed as medians and interquartile ranges. Asterisk indicates p = 0.0019 for comparison of group I with group III. Number sign indicates p = 0.0469 for comparison of group I with group III.
depression of the button before onset of sedation and analgesia. This was demonstrated in the 31% rate of unsuccessful attempts to obtain medication in the PCAS group. A pause in the advancement of the colonoscope for a few seconds each time the patient depresses the button may be a simple way to compensate for this delay. PCAS requires that a patient administer the drug in response to a stimulus, usually pain. In contrast, nurse administration of medication ideally prevents discomfort from occurring. Interestingly, the pain scores reported by the patients were similar for all groups, regardless of whether medication delivery was patient-controlled, through continuous infusion, or nurse-administered (Fig. 1). Despite comparable pain scores, patients allocated to the PCAS regimen were more satisfied with their degree of sedation than patients who received the continuous infusion of propofol/alfentanil or nurse-administered midazolam/meperidine. This corresponds to the results of a previous randomized controlled trial of PCAS with propofol/ alfentanil, which included satisfaction scores that were better than pain scores in the PCAS group.14 In view of the blinded design of our study, the high GASTROINTESTINAL ENDOSCOPY
5
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
level of satisfaction in the PCAS group indicates that the beneficial effect of patient-controlled drug administration during colonoscopy goes beyond the placebo effect of the PCAS equipment. The advantage of PCAS with propofol/alfentanil in terms of patient satisfaction is even more impressive considering that twice as many patients in the midazolam/meperidine group had amnesia compared with the patients who received propofol/alfentanil. This observation suggests that patients prefer to remember the colonoscopy and its self-controlled level of discomfort than to experience amnesia for the procedure. The use of transcutaneous carbon dioxide monitoring during endoscopy allows for more effective recognition of hypoventilation compared with standard clinical monitoring and the use of pulse oximetry alone.17 By using this method of monitoring, hypoventilation was found to be least in the PCAS group (Fig. 2). This safety advantage of PCAS may derive from the fact that pain usually precedes selfadministration of medication. Pain stimulates the cardiovascular and respiratory systems, which reduces the depressing effects of sedative and narcotic drugs. Furthermore, the PCAS method requires that the patient be conscious enough to depress the hand-held button. Consequently, oversedation is unlikely if rapidly acting drugs like propofol and alfentanil are used. In contrast, continuous infusion or nurse-administered dosing can deliver sedative medication when this is not actually needed. The more accentuated decrease in mean arterial blood pressure during the recovery period in the midazolam/meperidine group may be explained by the longer half-lives of these agents, which thereby results in persistent cardiovascular depression when the pain stimulus is no longer present. The faster recovery of patients and fewer after-effects in the propofol/alfentanil PCAS group compared with the midazolam/meperidine combination (Fig. 3) confirms published experience14 and is presumably due to the shorter half-lives of these drugs. The use of propofol in some countries requires the presence of an anesthetist, which may not be realistic in most endoscopy units. Safety aspects such as the need for assisted ventilation in cases of apnea, which has been reported after a 20-mg dose of propofol given during ERCP,22 are the reason for this recommendation. However, our data suggest that the application of propofol and alfentanil in incremental patientcontrolled bolus doses for colonoscopy may yield even better levels of safety compared with the conventional administration of midazolam and meperidine and, moreover, this results in a higher level of patient satisfaction and a shorter recovery period. 6
GASTROINTESTINAL ENDOSCOPY
ACKNOWLEDGEMENT We thank the Laubscher Company, Hölstein, Switzerland for providing the Graseby PCA Pump and the HP Viridia Patient-Monitor. REFERENCES 1. Froehlich F, Thorens J, Schwizer W, Preisig M, Kohler M, Hays RD, et al. Sedation and analgesia for colonoscopy: patient tolerance, pain and cardiorespiratory parameters. Gastrointest Endosc 1997;45:1-9. 2. Arrowsmith JB, Gerstman BB, Fleischer DE, Benjamin SB. Results from the American Society for Gastrointestinal Endoscopy/U.S. Food and Drug Administration collaborative study on complication rates and drug use during gastrointestinal endoscopy. Gastrointest Endosc 1991;37:421-7. 3. Jowell PS, Eisen G, Onken J, Bute BP, Ginsberg B. Patientcontrolled analgesia for conscious sedation during endoscopic retrograde cholangiopancreatography: a randomized controlled trial. Gastrointest Endosc 1996;43:490-4. 4. Stermer E, Gaitini L, Yudashkin M, Essaian G, Tamir A. Patient-controlled analgesia for conscious sedation during colonoscopy: a randomized controlled study. Gastrointest Endosc 2000;51:278-81. 5. White PF. Use of patient-controlled analgesia for management of acute pain. JAMA 1988;259:243-7. 6. Smythe M. Patient-controlled analgesia: a review [review]. Pharmacotherapy 1992;12:132-43. 7. Lehmann KA. New developments in patient-controlled postoperative analgesia [review]. Ann Med 1995;27:271-82. 8. Ganapathy S, Herrick IA, Geib AW, Kirkby J. Propofol patient-controlled sedation during hip or knee arthroplasty in elderly patients. Can J Anaesth 1997;44:385-9. 9. Hamid SK, McCann N, McArdle L, Asbury AJ. Comparison of patient-controlled sedation with either methohexitone or propofol. Br J Anaesth 1996;77:727-30. 10. Osborne GA, Rudkin GE, Jarvis DA, Young IG, Barlow J, Leppard PI. Intraoperative patient-controlled sedation and patient attitude to control. A crossover comparison of patient preference for patient-controlled propofol and propofol by continuous infusion. Anaesthesia 1994;49:287-92. 11. Dell RG, Cloote AH. Patient-controlled sedation during transvaginal cocyte retrieval: an assessment of patient acceptance of patient-controlled sedation using a mixture of propofol and alfentanil. Eur J Anaesthesiol 1998;15:210-5. 12. Heiman OR, Tolliver BA, Weis FR, O’Brien BL, DiPalma JA. Patient-controlled anesthesia for colonoscopy using propofol: results of a pilot study. South Med J 1998;91:560-4. 13. Roseveare C, Seavell C, Patel P, Criswell J, Shepherd H. Patient-controlled sedation with propofol and alfentanil during colonoscopy: a pilot study. Endoscopy 1998;30:482-3. 14. Roseveare C, Seavell C, Patel P, Criswell J, Kimble J, Jones C, et al. Patient-controlled sedation and analgesia, using propofol and alfentanil, during colonoscopy: a prospective randomized controlled trial. Endoscopy 1998;30:768-73. 15. Owens WD, Felts FA, Spitznagel EL. ASA physical status classification. A study of consistency of ratings. Anesthesiology 1978;49:239-43. 16. Morant J, Ruppanner H, editors. Swiss Compendium of Drugs. Basel, Switzerland: Documed; 2000. 17. Nelson DB, Freeman ML, Silvis SE, Cass OW, Yakshe PN, Vennes J, et al. A randomized, controlled trial of transcutaneous carbon dioxide monitoring during ERCP. Gastrointest Endosc 2000;51:288-95. 18. Trieger N, Newman MG, Miller JC. Measuring recovery from anesthesia—a simple test. Anesth Analg 1969;48:136-40. 19. Jung M, Hofmann C, Kiesslich R, Brackertz A. Improved VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
sedation in diagnostic and therapeutic ERCP: propofol is an alternative to midazolam. Endoscopy 2000;32:233-8. 20. Carlsson U, Grattidge P. Sedation for upper gastrointestinal endoscopy: a comparative study of propofol and midazolam. Endoscopy 1995;27:240-3. 21. Reimann FM, Samson U, Derad I, Fuchs M, Schiefer B,
D Külling, A Fantin, P Biro, et al.
Stange EF. Synergistic sedation with low-dose midazolam and propofol for colonoscopies. Endoscopy 2000;32:239-44. 22. Wehrmann T, Kokabpick S, Lembcke B, Caspary WF, Seifert H. Efficacy and safety of intravenous propofol sedation during routine ERCP: a prospective, controlled study. Gastrointest Endosc 1999;49:677-83.
Availability of Journal back issues As a service to our subscribers, copies of back issues of Gastrointestinal Endoscopy for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Customer Service, 6277 Sea Harbor Dr., Orlando, FL 32887 or call 800654-2452 or 407-345-4000 for information on availability of particular issues and prices. If unavailable from the publisher, photocopies of complete issues may be purchased from Bell & Howell Information and Learning, 300 N. Zeeb Rd., Ann Arbor, MI 48106; 313-761-4700.
VOLUME 54, NO. 1, 2001
GASTROINTESTINAL ENDOSCOPY
7
Background: The aim of this study was to assess the efficacy of patient-controlled analgesia and sedation with propofol/alfentanil for colonoscopy compared with continuous drug infusion and conventional nurse-administered medication. Methods: One hundred fifty patients undergoing colonoscopy on an outpatient basis were randomly assigned to 1 of 3 medication regimens. To maintain blinding, all patients were connected to an infusion pump. Group I patients could self-administer boluses of 4.8 mg propofol and 125 µg alfentanil without restriction. Group II patients received a continuous infusion with 0.048 mg/kg propofol and 0.12 µg/kg alfentanil per minute. Group III patients received intravenous premedication with 0.035 mg/kg midazolam and 0.35 mg/kg meperidine. Results: There were no differences between the groups with respect to pain (visual analogue scale) and procedure time. Patient-controlled analgesia and sedation with propofol/alfentanil (group I) resulted in less of an increase in the transcutaneous partial pressure of carbon dioxide (p = 0.0004) during colonoscopy and less of a decrease in mean arterial blood pressure (p = 0.0021) during recovery, as well as more complete recovery (p = 0.0019) after 45 minutes compared with conventional administration of midazolam/meperidine. Furthermore, patient-controlled analgesia and sedation yielded a higher degree of patient satisfaction than continuous infusion of propofol/alfentanil (p = 0.0033) or nurse-administered midazolam/meperidine (p = 0.0094). Conclusions: Patient-controlled administration of propofol and alfentanil for colonoscopy may provide a better margin of safety than conventional administration of midazolam and meperidine and results in a higher level of patient satisfaction and shorter recovery. (Gastrointest Endosc 2001;54:1-7.)
Various drugs are used to achieve sedation and analgesia during colonoscopy. Most endoscopists use sedative hypnotics such as midazolam or diazepam, alone or in combination with opiate narcotics, usually meperidine.1 Although some patients tolerate colonoscopy without sedation, most experience some discomfort, which may influence the successful completion of the examination. Variations in pain threshold, drug tolerance, and colonic sensitivity among individual patients make it difficult to anticipate the required dose of sedative medication. As a consequence, some patients are given an inadequate dose, whereas others are oversedated with the associated risk of cardiorespiratory complications and delayed recovery.2 Furthermore, patient versus physician assessment of discomfort during proceReceived November 28, 2000. For revision January 31, 2001. Accepted April 3, 2001. From the Division of Gastroenterology, Department of Internal Medicine, and the Institute of Anesthesiology, University Hospital of Zürich, Switzerland. Reprint requests: Daniel Külling, MD, Division of Gastroenterology, Department of Internal Medicine, University Hospital, Rämistr. 100, CH-8091 Zürich, Switzerland. Copyright © 2001 by the American Society for Gastrointestinal Endoscopy 0016-5107/2001/$35.00 + 0 37/1/116174 doi:10.1067/mge.2001.116174 VOLUME 54, NO. 1, 2001
dures differs significantly.3,4 To overcome the problem of individual differences in the perception of pain, patient-controlled analgesia (PCA) delivery systems were introduced as early as 1968. Since then, the results of many trials support the safety and efficacy of PCA for various pain syndromes including those relating to surgery, cancer, trauma, and obstetric procedures.5 Reviews of PCA indicate that patient acceptance is generally enthusiastic because of the possibility of self-control.6,7 The hypnotic and sedative agent propofol (2,6diisopropylphenol) and the opiate alfentanil both have rapid onset and termination of action, making them ideal choices for patient-controlled administration. Patient-controlled analgesia and sedation (PCAS) with propofol alone or in combination with alfentanil has been used during hip and knee arthroplasty, oral surgery, transvaginal oocyte retrieval, and also colonoscopy.8-14 These studies concluded that PCAS with propofol/alfentanil provides safe and satisfactory sedation and shortens recovery time. The present randomized doubleblind controlled trial compares PCAS with propofol and alfentanil with nurse-administered midazolam and meperidine and with continuous infusion of propofol and alfentanil without patient control in patients undergoing colonoscopy. GASTROINTESTINAL ENDOSCOPY
1
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
Table 1. Study design Group
I
Initial bolus Pump content Pump setting Additional boluses
Placebo (saline solution) Propofol + alfentanil PCAS Propofol + alfentanil
MATERIALS AND METHODS Patients and study design One hundred fifty consecutive patients scheduled for colonoscopy on an outpatient basis were enrolled in the study. Indications for colonoscopy included abdominal pain (39%), rectal bleeding or anemia (23%), altered bowel habits (17%), follow-up after polypectomy (16%), and miscellaneous (5%). Exclusion criteria were age below 18 or above 75 years, a history of colonic resection, allergy to study medication, severe cardiorespiratory disease (ASA classification III or higher),15 renal or hepatic impairment, seizure disorder, psychiatric disease, substance abuse, pregnancy or breast-feeding, and refusal to receive any medication for the forthcoming colonoscopy. All colonoscopies were performed by expert endoscopists who possessed equal levels of skill. According to a randomization list, patients were assigned to PCAS with propofol and alfentanil (group I), continuous infusion of propofol and alfentanil (group II), or conventional sedation with midazolam and meperidine (group III) (Table 1). Patients as well as endoscopists were blinded as to the sedation regimen being used. However, endoscopists may have been aware of patient attempts to depress the PCAS button to control discomfort, which could have biased endoscopist requests for additional boluses of medication. Patient information, preparation of the study medication, and patient monitoring during and after colonoscopy were performed by a gastroenterologist and/or a nurse who were not involved in the procedure. The study personnel were trained in cardiopulmonary resuscitation. An anesthetist was not present in the procedure room, but anesthesia support was on call at all times. The study was approved by the ethics committee of our hospital. Written informed consent was obtained. During this process, the education level of the patient was recorded and patients were asked to rate their anxiety in relation to the expected colonoscopy on a 10-cm visual analogue scale (VAS) (left end point “no anxiety,” right end point “very anxious”). Medication To maintain blinding, all patients received an intravenous bolus injection at the beginning of the colonoscopy (saline solution in group I and II, midazolam and meperidine in group III) (Table 1). Furthermore, all patients were connected to a PCA pump. They were instructed to depress the button as often as required to relieve discomfort. For patients in group I and II, the pump contained propofol and alfentanil; for group III it contained saline solution. In group I, the pump was programmed in PCAS mode, in group II and III for constant continuous infusion. 2
GASTROINTESTINAL ENDOSCOPY
II
III
Placebo (saline solution) Propofol + alfentanil Continuous infusion Propofol + alfentanil
Midazolam + meperidine Placebo (saline solution) Continuous infusion Midazolam or meperidine
As regulated by the manufacturers, propofol (10 mg/mL, Ansiven, Abbott, Baar, Switzerland) was mixed with alfentanil (0.5 mg/mL, Rapifen, Janssen-Cilag, Baar, Switzerland) at a volume-ratio of 20:1.16 In group I, the PCA pump (Graseby 3300, Graseby Medical, Watford, UK) was programmed to deliver a bolus dose of 0.5 mL (4.8 mg propofol and 12 µg alfentanil) and with a zero lockout interval.9,11,13,14 In group II, the pump was programmed at a constant continuous infusion rate of 0.005 mL/min × body weight in kg.11,15 In groups I and II, the study-nurse could administer additional 1-mL boluses whenever the endoscopist felt that the patient was uncomfortable and required additional sedation up to a maximal total dosage of 30 mL (288 mg propofol and 0.72 mg alfentanil). In group III, midazolam (1 mg/mL, Dormicum, Roche Pharma, Reinach, Switzerland) 0.035 mg/kg body weight and meperidine (50 mg/mL, Pethidin, Streuli, Uznach, Switzerland) 0.35 mg/kg body weight were administered intravenously before the start of the colonoscopy.1,2 During the procedure, alternating boluses of midazolam 0.015 mg/kg or meperidine 0.35 mg/kg were given as needed up to a maximal total dosage of 0.08 mg/kg midazolam and 0.7 mg/kg meperidine. Assessment The primary study objective was patient satisfaction with the degree of sedation during colonoscopy. Secondary parameters included pain, procedure time, amnesia, safety, and recovery. Ninety minutes after the end of the colonoscopy, the patients assessed their pain during the colonoscopy on a 10-cm VAS (left end point “no pain,” right end point “severe pain”) and their satisfaction with the degree of sedation (left end point “ideal sedation,” right end point “completely inadequate sedation”). Still without knowledge of the drug regimen they had received, the patients were also questioned as to whether they would request the same method of analgesia and sedation for a future colonoscopy (yes or no). In addition, the patients were given a questionnaire to be filled out the next day that asked about after-effects severe enough to disrupt their regular activities. Amnesia was tested by questioning the patients 90 minutes after the end of the procedure about the word (“peach”) they had been told to memorize during the colonoscopy and before withdrawal of the colonoscope. Patients were monitored during colonoscopy and recovery by clinical observation, continuous pulse oximetry, and automated sphygmomanometry at 5- and 15-minute intervals, respectively. All patients were given oxygen 2 L/min by nasal cannula during the entire procedure. During recovery, nasal oxygen supplementation was given VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
D Külling, A Fantin, P Biro, et al.
Table 2. Patient characteristics Group
I
Age (y) Male/female College degree Previous colonoscopies with/without sedation Anxiety before colonoscopy (10 cm VAS score)
II
55 (43-63) 56%/44% 22% 22%/18% 2.0 (0.0-5.0)
55 (40-65) 44%/56% 16% 30%/24% 3.0 (1.0-5.0)
III 48 (35-64) 54%/46% 24% 20%/24% 3.0 (0.0-5.0)
Data are expressed as medians (interquartile ranges) or percentages. There are no significant differences among the 3 groups.
for desaturation to less than 90%. Furthermore, patients were continuously monitored during colonoscopy by a transcutaneous carbon dioxide tension sensor (M1205A Viridia, Hewlett Packard, Andover, Mass.).17 Recovery from sedation was judged by using the Trieger dot-joining test before (baseline score), as well as 15, 45, and 90 minutes after colonoscopy. The Trieger test, which was adapted from the Bender Motor Gestalt test, measures sensory motor performance, a critical determinant of recovery. This test was first prospectively evaluated in 1969 and proved to be a simple reliable method to assess recovery after intravenous administration of sedative drugs or general anesthesia.18 The Trieger test has since been applied in several trials, including 2 recent endoscopic studies.12,19 Forty-two dots at 10- to 15-mm intervals are placed in the form of a 3-sided figure and a transversely placed bellshaped curve, with the 2 figures sharing 1 dot. The patients were asked to connect the dots with a pen. Scoring was calculated by multiplying the number of missed dots with the total error magnitude in millimeters. Statistical analysis and power calculation The medians and interquartile ranges (IQR) were calculated for quantitative variables. Mann-Whitney U rank sum tests were applied to compare the results from the 3 patient groups. Qualitative data were compared with Chisquare tests including Yates’ correction for continuity. A p value of less than 0.05 was considered significant. Power calculations were based on the primary objective of the study, which was patient satisfaction with the degree and quality of sedation. It was determined that 50 patients were needed in each group to detect a difference of 2 units or greater with a level of significance of 0.05, a power of 80%, and an expected variation of 3 units.
RESULTS Patients There were 77 men and 73 women, median age of 54 years (IQR 39-63), enrolled in the study. Another 16 patients who fulfilled entry criteria declined to participate. The 3 patient groups were similar with respect to age, gender, education level, previous colonoscopies with or without sedation, and precolonoscopy anxiety (Table 2). Medication Patients in the PCAS group (I) depressed the handheld button a median of 24 (IQR 16-45) times VOLUME 54, NO. 1, 2001
with 69% (IQR 53%-89%) being successful attempts that actually delivered 0.5-mL boluses. There was no difference with regard to the total number of attempts between groups II (23, IQR 12-51) and III (17, IQR 7-36). Both of the propofol/alfentanil groups received similar total drug doses. The patients in PCAS group I self-administered a median of 8.25 mL (IQR 6.0-12.5) or 78 mg (IQR 57-119) propofol and 198 µg (IQR 144-300) alfentanil, whereas group II received 9.45 mL (IQR 7.2-13.8) (equivalent to propofol 90 mg, IQR 68-131 and alfentanil 227 µg, IQR 173-331) by continuous infusion. This amount included one to four 1-mL nurseadministered boluses in 36% of group II patients, whereas no nurse-administered boluses were given to any group I patient. The median dosages of midazolam and meperidine in group III were 2.7 mg (IQR 2.3-3.0) and 27 mg (IQR 23-30), respectively. This amount included 1 or 2 additional boluses of medication during the procedure in 20% of group III patients. Satisfaction and amnesia Median procedure duration ranged from 22 to 25 minutes without significant differences between the 3 groups. Total colonoscopy with ileal or cecal intubation was achieved in all but 1 patient of group I, this procedure being abandoned in the sigmoid colon because of stenosing diverticulosis. All patients reported similar median pain scores (group I: 4.0, IQR 2.0-5.0; group II: 4.5, IQR 2.5-6.0; group III: 3.0, IQR 1.0-5.0) (Fig. 1). A higher percentage of patients medicated with midazolam/meperidine (50%) demonstrated amnesia for the period of the colonoscopy compared with patients who received propofol/alfentanil (24% in group I, p = 0.0129; 30% in group II, p > 0.05). Still, patients in the PCAS group (I) were more satisfied with the degree of sedation (10, IQR 9.0-10.0) than those in group II (8.5, IQR 6.0-10.0, p = 0.0033) and group III (8.5, IQR 6.510.0, p = 0.0094). Two patients in group III were not satisfied because the degree of sedation was too profound. Otherwise, all dissatisfied patients complained of inadequate sedation. At subgroup analysis, gender, education level, number of prior colonoscopies, indication for colonoscopy, and preGASTROINTESTINAL ENDOSCOPY
3
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
Figure 1. Visual analogue scores for pain during colonoscopy and patient satisfaction with the degree of sedation. Data are expressed as medians and interquartile ranges. Asterisk indicates p = 0.0033 for comparison of group I with group II; p = 0.0094 for comparison of group I with group III.
colonoscopy anxiety did not significantly influence the satisfaction advantage of PCAS. Still blinded as to drug regimen, more patients in group I (94%) than in group II (78%, p = 0.0437) and III (74%, p = 0.0141) indicated they would request the same method of sedation again in the event that a colonoscopy would be required in the future. Safety There were no complications related to the colonoscopy procedures. No serious respiratory or hemodynamic complications occurred with any drug regimen. During colonoscopy, there was no difference among the 3 groups with respect to changes in mean arterial blood pressure (Fig. 2) or oxygen saturation. Two patients in group I, 1 in group II, and none in group III had oxygen saturations between 88% and 90% with prompt normalization after increasing nasal oxygen supplementation from 2 to 4 L/min. PCAS patients exhibited less of an increase in transcutaneous partial pressure of carbon dioxide (pCO2) values (2.6 mm Hg, IQR 0.7-4.5) compared with groups II (4.1 mm Hg, IQR 3.0-8.3, p = 0.0120) and III (5.3 mm Hg, IQR 3.0-8.3, p = 0.0004). The changes in pulse rate and oxygen saturation during recovery were similar for all 3 groups. One patient in group I, 2 in group II, and 1 in group III required nasal oxygen delivery because of desatura4
GASTROINTESTINAL ENDOSCOPY
Figure 2. Safety. Increase in transcutaneous pCO2 during colonoscopy and decrease in mean arterial blood pressure (MABP) during colonoscopy and recovery. Data are expressed in mm Hg as medians and interquartile ranges. Asterisk indicates p = 0.0120 for comparison of group I with group II; p = 0.0004 for comparison of group I with group II. Number sign indicates p = 0.0021 for comparison of group I with group III.
tion to less than 90% with the lowest level being 86%. The patients who received midazolam/meperidine had a more accentuated decrease in mean blood pressure (8.5 mm Hg, IQR –1.0-15.0) than patients in the PCAS group (3.5 mm Hg, IQR –3.0-7.0, p = 0.0021). Recovery Fifteen minutes after the procedure, there was no difference between the 3 groups with regard to recovery from sedation as assessed by the Trieger dot-joining test (Fig. 3). However, at 45 and at 90 minutes, respectively, the Trieger scores of patients in the PCAS group were closer to baseline (0, IQR –0.5-0.5 and 0, IQR –1.0-0.5) than the scores of the patients who received midazolam/meperidine (1.0, IQR 0.0-9.0, p = 0.0019 and 0.25, IQR 0.0-1.5, p = 0.0469). Furthermore, more group III patients (36%) complained of after-effects during the 24 hours after the procedure than group I patients (11%, p = 0.006). The recovery scores of group II patients were VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
D Külling, A Fantin, P Biro, et al.
in between the results for group I and III patients without reaching statistical significance with respect to either group. DISCUSSION The results of this trial indicate that patientcontrolled analgesia and sedation with propofol and alfentanil during colonoscopy may result in better cardiorespiratory function and more rapid recovery compared with conventional nurse-administered midazolam and meperidine. Furthermore, PCAS with propofol/alfentanil yielded a higher degree of patient satisfaction than continuous infusion of the same drugs or nurse-administered midazolam/meperidine. The results of 3 randomized controlled endoscopic trials of PCAS are available, 2 with colonoscopy4,14 and 1 ERCP.3 In the studies by Jowell et al.3 and Stermer et al.,4 baseline sedation was induced with midazolam and meperidine in the “PCAS” study group with drug dosages that exceed the total doses given to our group III patients. Furthermore, the lockout interval (minimum time between doses) of the PCAS pump containing midazolam and meperidine was set to 5 minutes. Under these circumstances, it is unrealistic to assume a patient-controlled setting for a procedure such as colonoscopy in which the painful insertion phase usually lasts only 5 to 20 minutes. Therefore, it is not surprising that these 2 studies did not demonstrate any benefit of PCAS compared with standard analgesia and sedation.3,4 Similar to the present study, Roseveare et al.14 demonstrated that sole patient-controlled application of propofol and alfentanil during colonoscopy provided greater safety than conventional premedication with meperidine and diazepam. Propofol is at least as effective as midazolam for sedation in GI endoscopy and has a considerably faster onset and termination of action.19-22 A pilot study demonstrated that patient-controlled administration of propofol alone did not provide adequate pain relief for colonoscopy, but the addition of the opiate, alfentanil, offered adequate pain control and patient tolerance for the procedure.12 Based on published data,13,14 the lockout time of the PCAS pump was set at zero in the present study. Because the pump delivered the drug at a rate of approximately 200 mL/h, a 0.5-mL bolus (containing 4.8 mg of propofol and 12 µg alfentanil) was administered every 9 seconds if the hand-held button was repeatedly pressed. For comparison, single-induction doses for general anesthesia are 140 mg of propofol and up to 8.5 mg of alfentanil for a 70-kg patient.16 Even with the fast delivery system and rapid action of the drugs, there is necessarily a short delay after VOLUME 54, NO. 1, 2001
Figure 3. Recovery. Trieger scores above baseline 15, 45, and 90 minutes after colonoscopy. Data are expressed as medians and interquartile ranges. Asterisk indicates p = 0.0019 for comparison of group I with group III. Number sign indicates p = 0.0469 for comparison of group I with group III.
depression of the button before onset of sedation and analgesia. This was demonstrated in the 31% rate of unsuccessful attempts to obtain medication in the PCAS group. A pause in the advancement of the colonoscope for a few seconds each time the patient depresses the button may be a simple way to compensate for this delay. PCAS requires that a patient administer the drug in response to a stimulus, usually pain. In contrast, nurse administration of medication ideally prevents discomfort from occurring. Interestingly, the pain scores reported by the patients were similar for all groups, regardless of whether medication delivery was patient-controlled, through continuous infusion, or nurse-administered (Fig. 1). Despite comparable pain scores, patients allocated to the PCAS regimen were more satisfied with their degree of sedation than patients who received the continuous infusion of propofol/alfentanil or nurse-administered midazolam/meperidine. This corresponds to the results of a previous randomized controlled trial of PCAS with propofol/ alfentanil, which included satisfaction scores that were better than pain scores in the PCAS group.14 In view of the blinded design of our study, the high GASTROINTESTINAL ENDOSCOPY
5
D Külling, A Fantin, P Biro, et al.
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
level of satisfaction in the PCAS group indicates that the beneficial effect of patient-controlled drug administration during colonoscopy goes beyond the placebo effect of the PCAS equipment. The advantage of PCAS with propofol/alfentanil in terms of patient satisfaction is even more impressive considering that twice as many patients in the midazolam/meperidine group had amnesia compared with the patients who received propofol/alfentanil. This observation suggests that patients prefer to remember the colonoscopy and its self-controlled level of discomfort than to experience amnesia for the procedure. The use of transcutaneous carbon dioxide monitoring during endoscopy allows for more effective recognition of hypoventilation compared with standard clinical monitoring and the use of pulse oximetry alone.17 By using this method of monitoring, hypoventilation was found to be least in the PCAS group (Fig. 2). This safety advantage of PCAS may derive from the fact that pain usually precedes selfadministration of medication. Pain stimulates the cardiovascular and respiratory systems, which reduces the depressing effects of sedative and narcotic drugs. Furthermore, the PCAS method requires that the patient be conscious enough to depress the hand-held button. Consequently, oversedation is unlikely if rapidly acting drugs like propofol and alfentanil are used. In contrast, continuous infusion or nurse-administered dosing can deliver sedative medication when this is not actually needed. The more accentuated decrease in mean arterial blood pressure during the recovery period in the midazolam/meperidine group may be explained by the longer half-lives of these agents, which thereby results in persistent cardiovascular depression when the pain stimulus is no longer present. The faster recovery of patients and fewer after-effects in the propofol/alfentanil PCAS group compared with the midazolam/meperidine combination (Fig. 3) confirms published experience14 and is presumably due to the shorter half-lives of these drugs. The use of propofol in some countries requires the presence of an anesthetist, which may not be realistic in most endoscopy units. Safety aspects such as the need for assisted ventilation in cases of apnea, which has been reported after a 20-mg dose of propofol given during ERCP,22 are the reason for this recommendation. However, our data suggest that the application of propofol and alfentanil in incremental patientcontrolled bolus doses for colonoscopy may yield even better levels of safety compared with the conventional administration of midazolam and meperidine and, moreover, this results in a higher level of patient satisfaction and a shorter recovery period. 6
GASTROINTESTINAL ENDOSCOPY
ACKNOWLEDGEMENT We thank the Laubscher Company, Hölstein, Switzerland for providing the Graseby PCA Pump and the HP Viridia Patient-Monitor. REFERENCES 1. Froehlich F, Thorens J, Schwizer W, Preisig M, Kohler M, Hays RD, et al. Sedation and analgesia for colonoscopy: patient tolerance, pain and cardiorespiratory parameters. Gastrointest Endosc 1997;45:1-9. 2. Arrowsmith JB, Gerstman BB, Fleischer DE, Benjamin SB. Results from the American Society for Gastrointestinal Endoscopy/U.S. Food and Drug Administration collaborative study on complication rates and drug use during gastrointestinal endoscopy. Gastrointest Endosc 1991;37:421-7. 3. Jowell PS, Eisen G, Onken J, Bute BP, Ginsberg B. Patientcontrolled analgesia for conscious sedation during endoscopic retrograde cholangiopancreatography: a randomized controlled trial. Gastrointest Endosc 1996;43:490-4. 4. Stermer E, Gaitini L, Yudashkin M, Essaian G, Tamir A. Patient-controlled analgesia for conscious sedation during colonoscopy: a randomized controlled study. Gastrointest Endosc 2000;51:278-81. 5. White PF. Use of patient-controlled analgesia for management of acute pain. JAMA 1988;259:243-7. 6. Smythe M. Patient-controlled analgesia: a review [review]. Pharmacotherapy 1992;12:132-43. 7. Lehmann KA. New developments in patient-controlled postoperative analgesia [review]. Ann Med 1995;27:271-82. 8. Ganapathy S, Herrick IA, Geib AW, Kirkby J. Propofol patient-controlled sedation during hip or knee arthroplasty in elderly patients. Can J Anaesth 1997;44:385-9. 9. Hamid SK, McCann N, McArdle L, Asbury AJ. Comparison of patient-controlled sedation with either methohexitone or propofol. Br J Anaesth 1996;77:727-30. 10. Osborne GA, Rudkin GE, Jarvis DA, Young IG, Barlow J, Leppard PI. Intraoperative patient-controlled sedation and patient attitude to control. A crossover comparison of patient preference for patient-controlled propofol and propofol by continuous infusion. Anaesthesia 1994;49:287-92. 11. Dell RG, Cloote AH. Patient-controlled sedation during transvaginal cocyte retrieval: an assessment of patient acceptance of patient-controlled sedation using a mixture of propofol and alfentanil. Eur J Anaesthesiol 1998;15:210-5. 12. Heiman OR, Tolliver BA, Weis FR, O’Brien BL, DiPalma JA. Patient-controlled anesthesia for colonoscopy using propofol: results of a pilot study. South Med J 1998;91:560-4. 13. Roseveare C, Seavell C, Patel P, Criswell J, Shepherd H. Patient-controlled sedation with propofol and alfentanil during colonoscopy: a pilot study. Endoscopy 1998;30:482-3. 14. Roseveare C, Seavell C, Patel P, Criswell J, Kimble J, Jones C, et al. Patient-controlled sedation and analgesia, using propofol and alfentanil, during colonoscopy: a prospective randomized controlled trial. Endoscopy 1998;30:768-73. 15. Owens WD, Felts FA, Spitznagel EL. ASA physical status classification. A study of consistency of ratings. Anesthesiology 1978;49:239-43. 16. Morant J, Ruppanner H, editors. Swiss Compendium of Drugs. Basel, Switzerland: Documed; 2000. 17. Nelson DB, Freeman ML, Silvis SE, Cass OW, Yakshe PN, Vennes J, et al. A randomized, controlled trial of transcutaneous carbon dioxide monitoring during ERCP. Gastrointest Endosc 2000;51:288-95. 18. Trieger N, Newman MG, Miller JC. Measuring recovery from anesthesia—a simple test. Anesth Analg 1969;48:136-40. 19. Jung M, Hofmann C, Kiesslich R, Brackertz A. Improved VOLUME 54, NO. 1, 2001
Colonoscopy: patient-controlled analgesia and sedation with propofol and alfentanil
sedation in diagnostic and therapeutic ERCP: propofol is an alternative to midazolam. Endoscopy 2000;32:233-8. 20. Carlsson U, Grattidge P. Sedation for upper gastrointestinal endoscopy: a comparative study of propofol and midazolam. Endoscopy 1995;27:240-3. 21. Reimann FM, Samson U, Derad I, Fuchs M, Schiefer B,
D Külling, A Fantin, P Biro, et al.
Stange EF. Synergistic sedation with low-dose midazolam and propofol for colonoscopies. Endoscopy 2000;32:239-44. 22. Wehrmann T, Kokabpick S, Lembcke B, Caspary WF, Seifert H. Efficacy and safety of intravenous propofol sedation during routine ERCP: a prospective, controlled study. Gastrointest Endosc 1999;49:677-83.
Availability of Journal back issues As a service to our subscribers, copies of back issues of Gastrointestinal Endoscopy for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Customer Service, 6277 Sea Harbor Dr., Orlando, FL 32887 or call 800654-2452 or 407-345-4000 for information on availability of particular issues and prices. If unavailable from the publisher, photocopies of complete issues may be purchased from Bell & Howell Information and Learning, 300 N. Zeeb Rd., Ann Arbor, MI 48106; 313-761-4700.
VOLUME 54, NO. 1, 2001
GASTROINTESTINAL ENDOSCOPY
7