Secondary Logo

Journal Logo

Original Paper

Transition to post-operative epidural or patient-controlled intravenous analgesia following total intravenous anaesthesia with remifentanil and propofol for abdominal surgery

Bowdle, T. A.*; Ready, L. B.; Kharasch, E. D.; Nichols, W. W.; Cox, K.

Author Information
European Journal of Anaesthesiology: July 1997 - Volume 14 - Issue 4 - p 374-379



Remifentanil is a novel 4-anilidopiperidine muopioid [1]. Rapid hydrolysis by red cell and tissue esterase [2] results in an elimination half-life of about 10 min. The time required for a 50% reduction in effective site concentration is predicted to be less than 5 min, and is independent of the duration of remifentanil administration [3–5]. Therefore, continuous infusion of remifentanil is required to sustain an effect. The pharmacokinetic profile of remifentanil should be well suited to total intravenous anaesthesia (TIVA), in combination with a sedative-hypnotic drug such as propofol. The ultrashort duration of action confers potential advantages and disadvantages. On the one hand, emergence from anaesthesia should be very rapid and predictable. On the other hand, sudden discontinuation of remifentanil at the end of a major surgical procedure may result in rapid disappearance of analgesia and the abrupt onset of post-operative pain. Thus, a transition to longer acting analgesics is needed in order to preserve analgesia. The purpose of this study was to examine the transition to epidural analgesia or intravenous patient controlled analgesia (PCA), following TIVA with remifentanil and propofol. A post-operative infusion of remifentanil was initiated at the time of extubation and continued for 30 min in the recovery room. Thereafter, a transition was made to either morphine PCA or epidural analgesia.


Pre-operative preparation

Patients were enrolled after giving informed consent to a protocol approved by the institute. The patients were participants in a multicentre study of TIVA with remifentanil and propofol; the intra-operative results from the multicentre study have been published previously [6]. This study of the transition from TIVA to several modalities of post-operative pain control was performed at our centre as a branch of the multicentre study. All of the our patients participating in the multicentre study also participated in our branch study. Male and female patients, 18 years of age or older, ASA physical status I–IV, having elective inpatient abdominal surgery were eligible. Otherwise eligible patients were excluded from the study based on the following criteria: (a) clinically significant and unstable medical conditions; (b) chronic use of benzodiazepines or clonidine or acute use within 12 h of surgery; (c) chronic opioid use or acute use within 12 h of surgery; (d) clinically significant drug abuse (e.g. alcohol or illicit drugs) (e) weight greater than 100% of ideal body weight and (f) pregnant or breast-feeding women. Women of child bearing potential underwent urine pregnancy testing on the day of surgery.

On arrival in the operating room holding area, an i.v. line was inserted and midazolam, 0.025 mg kg−1 was administered. For those patients electing to have epidural post-operative analgesia, a lumbar epidural catheter was placed with lignocaine skin anaesthesia. Selection of epidural analgesia was based on patient preference, and was not randomized. No local anaesthetics or opioids were injected into the epidural catheter pre-operatively or intra-operatively.

Maintenance of anaesthesia

Anaesthesia was induced with remifentanil, 1 μg kg−1 i.v. bolus, followed by propofol, 0.5 mg kg−1 i.v. bolus. Immediately after the initial remifentanil bolus, a remifentanil infusion was initiated at a rate of 0.5–1 μg kg−1 min−1, supplemented by a propofol infusion of 75 μg kg−1 min−1. No inhaled anaesthetics were administered. Five minutes after tracheal intubation, the remifentanil infusion was reduced by 50%. Subsequently, remifentanil was titrated between 0.125 and 4.0 μg kg−1 min−1, based on clinical signs of depth of anaesthesia. Propofol could also be varied between 50 and 125 μg kg−1 min−1. Increases in remifentanil or propofol infusion rates were preceeded by bolus doses of 1.0 μg kg−1 or 20 μg kg−1, respectively.

Vecuronium was titrated by intermittent boluses to maintain adequate surgical relaxation, with at least one twitch of the train-of-four (TOF) maintained at all times.

Extubation and recovery

Approximately 15 min prior to the end of surgery, the propofol infusion was reduced by 50%. Vecuronium was reversed with neostigmine and glycopyrrolate. At the time of the last suture or skin staple the propofol infusion was discontinued and the remifentanil infusion was reduced to 0.05 mg kg−1 min−1 (henceforth referred to as the 'post-operative infusion'). Patients were extubated when adequate spontaneous ventilation and response to verbal commands was established. The remifentanil post-operative infusion was continued for at least 45 min. During the first 30 min, the infusion was titrated with the goal of maintaining pain at a level of 0 or 1 on a scale of 0–3 (0 = no pain, 1 = mild pain, 2 = moderate pain, 3 = severe pain). Titration was accomplished with boluses of 0.25–0.5 μg kg−1, accompanied by infusion rate increases of 0.025–0.1 μg kg−1 min−1 (depending on the severity of the pain), with each change of dose seperated by at least 5 min.

After treating 13 patients, the multicentre protocol for post-operative remifentanil analgesia was modified because of inadequate pain control immediately following emergence from anaesthesia, reported in some of the centres of the multicentre trial. The initial remifentanil analgesic infusion was increased from 0.05–0.10 μg kg−1 min−1. Nine patients were treated under the modified post-operative infusion protocol. There were no significant differences between the original protocol patients and the modified protocol patients for any of the variables reported below, therefore the data from the two sets of patients have been combined.

Thirty minutes after initiating the remifentanil post-operative infusion, morphine, 0.05 mg kg−1 i.v., was given. Subsequently, 12 patients who elected to have PCA for post-operative analgesia received additional loading doses of morphine and commenced on PCA according to our standard Acute Pain Service protocol [7]. The typical PCA starting dose is 1 mg, with an 8 min lockout interval, increased to 1.5 mg with a 6 min lockout interval as required. Typical doses were modified as required by the Acute Pain Service based upon the requirements of individual patients. For the 10 patients who elected to have epidural analgesia, the epidural catheter was tested with lignocaine, 60 mg, and adrenaline, 15 μg. Lignocaine, 0.5% with adrenaline 1:200 000, was then administered in a dose sufficient to produce sensory block above the upper level of the incision. Epidural preservative free morphine was then administered according to our Acute Pain Service protocol [7]. The typical epidural morphine dose was a 2–5 mg bolus, depending on patient age and the type of surgery, repeated every 6–12 h. As for PCA, typical doses were modified as required by the Acute Pain Service. The remifentanil post-operative infusion was reduced by 50%, 15 min after the first dose of i.v. morphine, and was subsequently tapered and discontinued as epidural analgesia or PCA was provided. Remifentanil was discontinued within 60 min of the first dose of i.v. morphine for 19 of the 22 patients, and within 80 min for the remaining three patients.

Pain scores in the recovery room were assessed by the investigators using the 0–3 scale described above at 1, 3, 5, 7, and 10 min and subsequently every 5 min for 30 min from extubation until the initial morphine dose, then at 1, 3, 5, 7, 10, and 15 min and subsequently every 5 min from the initial morphine dose until the remifentanil was discontinued, then at 1, 3, 5, 10, 20, 30, 45, and 60 min and subsequently every 15 min from the time remifentanil was discontinued until discharge from the recovery room. Once the patient arrived on the surgical ward, pain scores at rest and with coughing were assessed by the Acute Pain Service [7] every 24 h using a scale of 0–10 (0 being no pain and 10 being the worst pain imaginable to the patient).

Statistical analysis

Statistical analysis consisted of Student's t-test for comparison of group means, z-test for comparison of proportions, and χ2-test for comparison of pain scores.


Twenty-two patients were enrolled, 18 females and four males (excluding four patients that did not complete the study or were studied in an unblinded fashion to familiarize the investigators with the protocol). The mean age was 50.9 (SD = 13.6) years and the mean weight was 74.8 (SD = 18.2) kg. The operations performed on women were major gynaecological oncology procedures. The operations performed in the male patients were a cholecystectomy, a radical prostatectomy, and two ventral hernia repairs. Patients awakened from anaesthesia promptly. The mean time to response to verbal command was less than 6 min after discontinuation of propofol. The mean time to extubation was less than 8 min following discontinuation of propofol.

During the first 30 min in the recovery room, analgesia was provided solely by the infusion of remifentanil. The median infusion rate at 30 min (just prior to the administration of the initial morphine dose) was 0.086 μg kg−1 min−1. The range was 0.04–0.26 μg kg−1 min−1. The rate of infusion was relatively stable. For 14 of the 22 patients studied, the infusion rate was not changed from the initial rate. For six patients (four in the 0.05 μg kg−1 min−1 starting dose group, and 2 in the 0.10 μg kg−1 min−1 group) the infusion rate was increased, and for two patients (one in each group) the infusion rate was decreased. Immediately prior to administration of morphine, all patients had pain scores ≤2, and 86% had pain scores of 0 or 1. Following administration of i.v. morphine, institution of PCA or epidural analgesia, and discontinuation of the remifentanil infusion, 95% of patients had pain scores of 0 or 1 (χ2; P = 0.24). Pain scores in the recovery room were not significantly different for patients with epidural analgesia compared with PCA (χ2; P=0.19).

Following discharge from the recovery room, epidural analgesia regimens consisted of morphine boluses, except for two patients who received infusions of morphine plus bupivacaine or fentanyl plus bupivacaine. Acute Pain Service scores were significantly lower for epidural analgesia compared with PCA, both for pain at rest (χ2; P = 0.034) and for pain with coughing (χ2; P=0.044) (Fig. 1).

Fig. 1.
Fig. 1.:
Pain scores on a scale of 0–10 were recorded daily by the Acute Pain Service until analgesic therapy was changed to oral medication. Pain scores were significantly lower for epidural analgesia (LEP) compared with PCA, both for pain at rest (χ2; P = 0.034) and for pain with coughing (χ2; P = 0.044).

There was no nausea or vomiting during the infusion of remifentanil, prior to the administration of morphine. Following administration of morphine, 16 of the 22 patients had nausea. The difference between the proportion of patients with nausea before morphine (0/22) and the proportion of patients with nausea after morphine administration (16/22) was significant (z-test; P<0.01). Of the patients with nausea, four had nausea prior to discontinuation of remifentanil, and 12 had nausea after the discontinuation of remifentanil, a significant difference (z-test; P<0.01). Emesis occurred in four patients. Following discharge from the recovery room, there was no significant difference in nausea between those patients receiving PCA and those receiving epidural analgesia, nor was there a difference in itching. No respiratory complications were identified in any patient.


Emergence from TIVA with remifentanil and propofol was very rapid, because of the rapid offset of both drugs. While rapid awakening is desirable, pain is likely to be a problem following major surgery if remifentanil levels are allowed to fall rapidly prior to the administration of a longer acting analgesic. The purpose of this study was to examine methods for providing analgesia following TIVA. A remifentanil analgesic infusion was used to smooth the transition between remifentanil intra-operative analgesia and either PCA morphine or epidural analgesia. The post-operative remifentanil infusion produced satisfactory analgesia, with mean pain scores in the 'mild' range. Titration of remifentanil was not difficult, with 14/22 patients remaining on the initial dose of 0.05 or 0.1 μg kg−1 min−1. Thus, it appears that the dosing range of 0.05–1.0 μg kg−1 min−1 represents a reasonable starting dose for the control of pain following major surgery. There was a five-fold range of remifentanil doses at the end of the 30 min titration period, consistent with other studies that have shown a wide range of opioid requirements in surgical patients [8]. Most of the patients underwent lower abdominal operations. Patients undergoing other types of surgery may have analgesic requirements that vary substantially from those reported here.

The transition from the post-operative remifentanil infusion to PCA or epidural analgesia was smooth. Pain scores during the transition were not significantly different from pain scores during the remifentanil infusion. Epidural analgesia was established initially with 0.5% lignocaine, followed immediately by epidural morphine. By the time the lignocaine had worn off, the epidural morphine had taken effect. The epidural catheter was not activated intra-operatively in order to observe the effects of the remifentanil intra-operatively [6] and the post-operative infusions, in the absence of epidural analgesia. In routine clinical practice the epidural catheter is more likely to be used intra-operatively as well as post-operatively, in which case epidural analgesia would already be established when the patient emerged from anaesthesia.

Pain scores were significantly lower for epidural analgesia compared with PCA morphine on the surgical ward, both at rest and with coughing. The patients were not randomized to epidural analgesia vs. PCA, and this study was not intended to be a rigorous comparison of these modalities. Numerous comparisons of epidural and i.v. routes of opioid administration for post-operative pain have been reported [9–12]. Taken as a whole, these studies found better analgesia with epidural opioids, generally without worse side effects. The results of the present study are consistent with this conclusion. However, some studies found no difference in the effects of epidural and i.v. opioid administration, particularly for more lipid soluble opioids that are rapidly absorbed from the epidural space into the systemic circulation [10, 13–15].

Opioids commonly produce nausea, and the incidence of nausea in this study is similar to that in previous reports involving administration of opioids [16]. The analgesic infusion of remifentanil by itself was not associated with any nausea. Nausea occurred only after administration of i.v. morphine. Whether the nausea was related to morphine, the combination of morphine and remifentanil, or declining remifentanil levels (remifentanil was tapered and discontinued after morphine administration), is unknown. Experiments in cats have suggested that opioids have naloxone-sensitive antiemetic effects in the vomiting centre of the brain stem, and emetic effects on the chemoreceptor trigger zone [17]. Whether opioids produce nausea and vomiting may depend upon the balance of the opposing effects. Fentanyl did not produce emesis in cats unless they were pretreated with i.v. naloxone; presumably the naloxone antagonizes the antiemetic effects of fentanyl in the vomiting centre, resulting in emesis [17]. Naloxone administration to patients pretreated with morphine has also been noted to produce emesis [18]. Perhaps, tapering the remifentanil infusion resulted in a sudden fall in remifentanil levels in the brain stem, altering the balance between emetic and antiemetic effects, and triggering the onset of nausea and vomiting. Additional studies may be useful to determine whether sudden reductions in remifentanil levels promote nausea and vomiting, as the present study suggests.

In conclusion, a post-operative remifentanil infusion was used to smooth the transition from TIVA with remifentanil and propofol to longer acting analgesics, such as PCA morphine and epidural analgesia. This method resulted in rapid emergence from anaesthesia while limiting pain. The rapid onset and offset of remifentanil make it well suited to titration of analgesic effects in the recovery room. However, management of continuous opioid infusions requires a high level of supervision, and may be beyond the scope of many recovery rooms. As an alternative, longer acting analgesics could be administered prior to emergence from anaesthesia. While remifentanil might be discontinued sooner if longer acting analgesics were given intra-operatively, emergence from anaesthesia might be slower, and titration of analgesia might be more difficult. Morphine in particular has a very slow onset of action that is not well suited to moment-to-moment titration of effect [19]. As additional experience is gained with the use of remifentanil for TIVA and balanced anaesthesia, it will be important to refine the methods described here and to devise other methods for the transition from intra-operative to post-operative analgesia.


This study was supported in part by Glaxo Inc. Research Institute, Research Triangle Park, NC, USA.


1 Feldman PL, James MK, Brackeen MF et al. Design, synthesis, and pharmacological evaluation of ultrashort- to long-acting opioid analgetics. J Med Chem 1991; 34: 2202–2208.
2 Selinger K, Nation RL, Smith GA. Enzymatic and chemical hydrolysis of remifentanil. Anesthesiology 1995; 83: A385.
3 Egan TD, Lemmens HJM, Fiset P et al. The pharmacokinetics of the new short-acting opioid remifentanil (GI897084B) in healthy adult male volunteers. Anesthesiology 1993; 79: 881–892.
4 Glass PSA, Hardman D, Kamiyama Y et al. Preliminary pharmacokinetics and pharmacodynamics of an ultrashort-acting opioid: remifentanil (GI87084B). Anesth Analg 1993; 77: 1031–1040.
5 Westmoreland CL, Hoke JF, Sebel PS, Hug CC, Muir KT. Pharmacokinetics of remifentanil (GI87084B) and its major metabolite (GI90291) in patients undergoing elective inpatient surgery. Anesthesiology 1993; 79: 893–903.
6 Hogue CW, Bowdle TA, O'Leary C et al. A multicenter evaluation of total intravenous anesthesia with remifentanil and propofol for elective inpatient surgery. Anesth Analg 1996; 83: 279–285.
7 Ready LB, Oden R, Chadwick HS et al. Development of an anesthesiology-based postoperative pain management service. Anesthesiology 1988; 68: 100–106.
8 Ausems ME, Vuyk J, Hug CC, Stanski DR. Comparison of a computer-assisted infusion vs. intermittent bolus administration of alfentanil as a supplement to nitrous oxide for lower abdominal surgery. Anesthesiology 1988; 68: 851–861.
9 Loper KA, Ready LB, Nessly M, Rapp S. Epidural morphine provides greater pain relief than patient-controlled intravenous morphine following cholecystectomy. Anesth Analg 1989; 69: 826–828.
10 Loper KA, Ready B, Downey M et al. Epidural and intravenous fentanyl infusions are clinically equivalent after knee surgery. Anesth Analg 1990; 70: 72–75.
11 Madej TH, Wheatley RG, Jackson JB, Hunter D. Hypoxaemia and pain relief after lower abdominal surgery: comparison of extradural and patient-controlled analgesia. Br J Anaesth 1992; 69: 554–557.
12 Egan KJ, Ready LB. Patient satisfaction with intravenous PCA or epidural morphine. Can J Anaesth 1994; 41: 6–11.
13 Glass PSA, Estok P, Ginsberg B, Goldberg JS, Sladen RN. Use of patient-controlled analgesia to compare the efficacy of epidural to intravenous fentanyl administration. Anesth Analg 1992; 74: 345–351.
14 Chauvin M, Hongnat JM, Mourgeon E, Lebrault C, Bellenfant F, Alfonsi P. Equivalence of postoperative analgesia with patient-controlled intravenous or epidural alfentanil. Anesth Analg 1993; 76: 1251–1258.
15 Liu S, Carpenter RL, Mulroy MF et al. Intravenous vs. epidural administration of hydromorphone: effects on analgesia and recovery after radical retropubic protatectomy. Anesthesiology 1995; 82: 682–688.
16 Watcha MF, White PF. Postoperative nausea and vomiting Anesthesiology 1992; 77: 162–184.
17 Costello D, Borison H. Naloxone antagonizes narcotic self-blockade of emesis in the cat. J Pharmacol Exp Ther 1977; 203: 222–230.
18 Longnecker D, Grazis P, Eggers G. Naloxone antagonism of morphine induced respiratory depression. Anesth Analg 1973; 52: 447–452.
19 Hug C, Morphy M, Rigel E, Olson W. Pharmacokinetics of morphine injected intravenously into anesthetized dog. Anesthesiology 1981; 54: 38–47.

Anaesthesia, intravenous, remifentanil, propofol; Analgesics, remifentanil; Pain, post-operative; Anaesthetic Techniques, epidural, patient-controlled analgesia

© 1997 European Academy of Anaesthesiology