Secondary Logo

Journal Logo

Original Article

Remifentanil compared with morphine for postoperative patient-controlled analgesia after major abdominal surgery: a randomized controlled trial

Kucukemre, F.1; Kunt, N.2; Kaygusuz, K.2; Kiliccioglu, F.2; Gurelik, B.2; Cetin, A.3

Author Information
European Journal of Anaesthesiology: May 2005 - Volume 22 - Issue 5 - p 378-385
doi: 10.1017/S0265021505000657


The goal of postoperative pain management after abdominal surgery is to relieve pain so that normal functions such as ventilation, gastrointestinal motility, coughing and mobility are only minimally impaired. Patient-controlled analgesia (PCA), which allows patients to self-administer frequent, small doses of opioids as needed to manage pain is an effective method for this purpose [1]. The most frequently used opioid in PCA is morphine [2] but many other opioids such as fentanyl, meperidine, and alfentanil have been tested [3]. There is no evidence that they offer any benefits over morphine [4]. The typical side-effects of opioids, such as respiratory depression, sedation, nausea, vomiting and pruritus may sometimes restrict the successful application of PCA. Ideally, the analgesic for PCA should have a rapid onset of action, be highly efficacious in relieving pain, have an intermediate duration of action (30-60 min), produce no tolerance or dependence and have no or minimal side-effects or adverse drug effects [5].

Remifentanil is a potent, selective μ-opioid receptor agonist with unique pharmacokinetic properties [6]. It has a rapid onset of peak effect, a short duration of action independent of the duration of infusion and rapid clearance (40 mL−1 kg−1 min−1) [7-9]. The unique pharmacokinetic profile of remifentanil is attributed to rapid metabolism of the propanoic methyl ester linkage on the parent piperidine molecule by non-specific esterases in blood and tissues [10].

This randomized, double-blinded clinical study was designed to compare the efficacy and safety of remifentanil and morphine in intravenous (i.v.) patient-controlled analgesia during the first 24 h following major abdominal surgery.


After obtaining the approval of the local Ethics Committee, written informed consent was obtained from patients undergoing major elective abdominal surgery. For sample size determination, an average difference of 15 mm on a 100-mm visual analogue scale (VAS) used for pain assessment was defined as clinically relevant. It was expected that 95% of the reported VAS scores would range between 0 and 80 mm, resulting in a standard deviation (SD) of 20 mm. A total sample size of 60 patients (30 in each arm of the study) was calculated to be necessary to detect a 15-mm reduction of VAS score with a power of 80% and an alpha error of 5%. It was assumed that the study drop-out rate would be approximately 15%, and therefore a total sample of 69 patients were recruited and randomly allocated to one of two groups. Randomization was achieved by using sequentially numbered, opaque-sealed envelopes containing computer-generated random allocations in a ratio of 1 : 1 in balanced blocks of eight.

We studied 69 adult American Society of Anesthesiologists (ASA) I-II patients undergoing elective major abdominal surgery through a midline incision. Patients under 18 yr or over 65 yr of age or with neurological, psychiatric, metabolic, endocrine, renal or hepatic disorders, or taking analgesic, sedative, antidepressant, antiepileptic, antiemetic or steroid drugs were excluded. Other criteria for exclusion included obesity (body mass index (BMI) > 30), refusal of consent, allergy to or contraindications for either of the study drugs, inability to use PCA, difficulty in communication, history of abdominal surgery, substance abuse, fever, shivering requiring treatment and infection.

The day before surgery, age, weight, gender and history of general anaesthesia were collected. All patients were taught how to use the PCA pump (Provide Pain Management®; Abbott Laboratories, North Chicago, IL, USA). Preoperative pulse and respiration rate, systolic and diastolic blood pressure (BP) and arterial blood gases were recorded. The intraoperative anaesthetic management was standardized in all patients. General anaesthesia was induced with thiopental (4-7 mg kg−1), vecuronium (0.1 mg kg−1) and fentanyl (1 μg kg−1). Anaesthesia was maintained with nitrous oxide (N2O) (66%) in oxygen (O2) and sevoflurane (1.5-3% expired concentration). It was not necessary to antagonize residual neuromuscular block in any patient. No other sedative, analgesic or antiemetic drug was administered. Type and duration of surgery was recorded in all patients.

In the post-anaesthesia care unit patients received standard postoperative care, including O2 administration via a nasal cannula (3 L min−1). Discharge criteria included being awake and oriented, able to breath deeply and cough freely, BP within 20% of preoperative values, temperature >36.0°C, absence of shivering, minimal pain and minimal nausea. Time spent (min) in the recovery unit was determined on a case-by-case basis and recorded.

In the recovery unit, collection of clinical data was begun as soon as the patient complained of pain (time zero) and after that it was continued on the ward for 24 h at the following time points (hours after operation) 1, 2, 4, 6, 12, 18 and 24. The clinical data were the pulse and respiration rates, systolic and diastolic BP, arterial blood gases (pH, PaO2 and PaCO2), oxygen saturation (SaO2), sedation and visual analogue scores and the presence of nausea and vomiting, shivering, and any other side-effects. The degree of sedation was evaluated with the Ramsay sedation scale [11]. Pain intensity was measured using a horizontal 100-mm VAS at rest and on movement. Patients indicated the severity of pain with a mark along the line between 0, no pain and 100, worst pain imaginable. The scale was explained to patients in the preoperative period. The severity of post-anaesthetic shivering was assessed according to a five-point scale similar to that used by Wrench and colleagues [12]. If a score ≥2 was reached, 25 mg meperidine was given i.v. as rescue medication and repeated once if necessary. These patients were excluded from further evaluation. The postoperative nausea and vomiting (PONV) score (0, no nausea; 1, mild nausea; 2, moderate nausea; 3, severe nausea and 4, severe nausea and vomiting) was recorded. Nausea lasting more than 10 min and vomiting were treated with i.v. ondansetron (4 mg) at 30-min intervals to a maximum dose of 8 mg.

After initial pain assessment, the patients in the remifentanil group received i.v. remifentanil PCA with a loading dose of 45 μg, a maintenance dose of 1 μg min−1, a bolus dose of 15 μg and a lockout interval of 10 min. Those in the morphine group received i.v. morphine PCA with a loading dose of 5 mg, a maintenance dose of 5 μg min−1, a bolus dose of 1 mg, and a lockout interval of 15 min. The purpose of these background infusions was to maintain therapeutic drug levels in the blood while the patients are not self-initiating boluses, such as during sleep.

Pulmonary function was tested before induction of anaesthesia as well as at 4 and 26 h after operation. Forced expiratory volume in 1 s (FEV1), forced expiratory vital capacity (FVC) and FEV1/FVC ratio were determined.

We assumed an equianalgesic bolus-dose ratio of remifentanil to morphine of 66 : 1. Since there was no data directly comparing the two opioids, we benefited from a study including information about the equianalgesic dose of fentanyl and morphine found in the existing literature [13]. For the equianalgesic doses of loading and maintenance, we benefited from a study including a comparison of remifentanil and morphine sulfate for acute postoperative analgesia [14]. Data on demand, delivery and total accumulated dose were retrieved from the PCA computer memory. The total drug use, number of bolus doses delivered, number of bolus doses demanded and the delivery/demand ratio were recorded. Pain relief was provided by the study drugs with PCA during the first 24 postoperative hours and after that intramuscular (i.m.) diclofenac sodium (75 mg) was administered as soon as PCA was terminated.

All patients were asked on the questionnaire to rate their overall pain experience (0, none; 1, mild; 2, moderate or 3, severe) and their degree of satisfaction with the management of their pain (0, poor; 1, adequate; 2, good; 3, excellent). Primary efficacy assessments included the proportion of patients with successful analgesia defined as no or mild pain with respiratory rate ≥11 and SaO2 ≥94% [15] for the entire period after administration of the study drugs and with an excellent degree of satisfaction.

Data were presented as mean ± SD or percentage as appropriate. Statistical analyses were performed with the programme package Statistica for Windows 6.0. (Statsoft, Inc., Tulsa, AR, USA). For pulse and respiratory rates, systolic and diastolic BP, arterial blood gases, SaO2, pulmonary function results, and VAS and sedation scores, one- or two-way repeated-measures analysis of variance (ANOVA) with Dunnett's test as post hoc test when appropriate were performed to test the effects of drug (between-group factor) and measurement time (within-group factor). We used t-tests to compare age, weight, duration of surgery, time spent in the recovery unit, total drug use, number of bolus doses demanded and delivered, delivery/demand ratio, and total PONV score. Sex, history of general anaesthesia, type of surgery, overall nausea/vomiting ratio, need for use of ondansetron, overall pain experience and patient's degree of satisfaction between the groups were compared with χ2. A P-value of <0.05 was considered significant.


During the study period, 91 consecutive patients with the required surgical indication were identified. Two had to be excluded because of obesity, seven refused to participate, and 13 were missed because of high workload. Recruitment ended with the completion of the protocol in the 69th patient. Of these 69 patients, four in the remifentanil group and five in the morphine group had to be excluded from data analysis because of protocol violations. One developed a wound haematoma requiring a surgical intervention (remifentanil group), two had incomplete data (one in each group), three required the use of meperidine for the treatment of shivering (two in the remifentanil and one in the morphine group), two were unable to use the PCA system (morphine group), and one had to undergo reoperation because of a surgical complication (morphine group).

There were no significant differences in age (46.6 ± 13.3 vs. 47.2 ± 12.8 yr), weight (72.2 ± 6.5 vs. 72.2 ± 6.5 kg), sex ratio (M : F) (6 : 24 vs. 7 : 23), history of general anaesthesia (4 : 26 vs. 5 : 25), duration of surgery (2.4 ± 0.8 vs. 2.2 ± 0.8 h) or time spent in the recovery unit (39.4 ± 16.7 vs. 45.9 ± 18.6 min). The type of surgery was as follows: cholecystectomy (15% vs. 18%), colectomy (21% vs. 19%), radical hysterectomy (22% vs. 28%) and laparotomy (42% vs. 35%). There was no significant difference with respect to the type of surgery.

Table 1 lists pH, PaO2 and PaCO2. Figure 1 shows systolic and diastolic BP and pulse rate, and Figure 2 gives respiratory rate and SaO2. There was no significant difference in these parameters except for the respiratory rate, which was significantly lower with morphine (P < 0.05). Respiratory and circulatory parameters changed significantly from baseline during the course of PCA in both groups as shown in Table 1 and Figures 1 and 2 (P < 0.05). However, no patient in any group required a therapeutic intervention with regard to haemodynamic and respiratory effects of study medications. No patient in either group became unrousable or had a respiratory rate <11 min−1 except for one patient who developed a short period of apnoea after the loading dose of remifentanil. We treated this with O2 and ventilatory support by mask for 5 min.

Table 1
Table 1:
pH, PaO2 and PaCO2 in remifentanil (R) and morphine (M) groups.
Figure 1.
Figure 1.:
Systolic and diastolic BP and pulse rates. Data were expressed as mean ± SD. Tp, preoperative; T0, postoperative 0 h; T1, postoperative 1 h; T2, postoperative 2 h; T4, postoperative 4 h; T6, postoperative 6 h; T12, postoperative 12 h; T18, postoperative 18 h; and T24, postoperative 24 h. aP < 0.05 vs. Tp in morphine group. bP < 0.05 vs. T0 and T2-T24 in remifentanil group. cP < 0.05 vs. T0-T24 in morphine group.
Figure 2.
Figure 2.:
Respiratory rate and SaO2. Data were expressed mean ± SD. For expansions for the groups names refer to caption forFigure 1. dP < 0.05 vs. T0-T24 in remifentanil group. eP < 0.05 vs. T0 and T12-T24 in morphine group. fP < 0.05 vs. T6 in morphine group.

No significant difference was seen between the sedation scores between the groups (Fig. 3) and there was no significant change in the sedation scores in the course of PCA.

Figure 3.
Figure 3.:
Sedation scores. Data were expressed as median (minimum and maximum). For expansions for these groups names refer to caption forFigure 1.

No significant differences were found in the VAS scores at rest and on movement between the groups (Fig. 4). The VAS scores at rest and on movement of 1 and 24 h were significantly lower than that at zero time in both groups (P < 0.05).

Figure 4.
Figure 4.:
Visual analogue scores at rest (a) and on movement (b). Data were expressed as median (minimum and maximum). For expansions for these groupsnames refer to caption forFigure 1. a,b,c,dP < 0.05 vs. T1-T24.

No significant differences in FEV1, FVC and FEV1/FVC ratios were found between the groups (Fig. 5). FEV1 preoperatively was significantly higher than that at 4 h in both groups (P < 0.05) but this was not considered to be a clinically important finding.

Figure 5.
Figure 5.:
FEV1, FVC and FEV1/FVC ratios. Data were expressed mean ± SD. Tp, preoperative; T4, postoperative 4 h; and T26, postoperative 26 h. a,bP < 0.05 vs. T4.

Table 2 gives total drug use, bolus delivery, bolus demand and delivery/demand ratio of the study groups. When their equianalgesic doses were compared, the total drug use in the remifentanil group was significantly higher than that of the morphine group (P < 0.05). The bolus delivery, bolus demand and delivery/demand ratio of the remifentanil group was significantly higher than that of the morphine group (P < 0.05).

Table 2
Table 2:
Drug usage data.

Between the groups, there were no significant differences in the ratio of overall presence of PONV (32% vs. 34%), need for the use of ondansetron (27% vs. 29%) and the ratio of shivering (9% vs. 6%). There was also no significant difference between groups with regard to total PONV score. None of the patient requested termination of PCA due to these side-effects. No rescue analgesic drug was administered during PCA use. With these dose settings for remifentanil and morphine, there was no significant finding of short-term tolerance.

Calculating the cost of the two agents based on their market price in our country and using a mean total dose of 2.3 mg for remifentanil and 31.0 mg for morphine we found that the drug costs of remifentanil was 40 times higher than that of morphine.

There were no significant differences in the ratio of successful analgesia (36.7% vs. 43.3%) and the excellent degree of satisfaction (26.7% vs. 39%) between the remifentanil and morphine groups, respectively. There was no significant difference in the ratio of patients with SaO2 <94% (0.12 (0.03-0.27) vs. 0.10 (0-0.23)) between remifentanil and morphine groups, respectively.


The results of the present study confirm that patient-controlled analgesia with i.v. remifentanil provides the same quality of analgesia and patient satisfaction after abdominal surgery as morphine in these clinical settings. Remifentanil caused a degree of overall sedation comparable to that with morphine. Remifentanil PCA with a loading dose of 45 μg, a maintenance dose of 1 μg min−1, a bolus dose of 15 μg and a lockout interval of 10 min can be suggested as an acceptable method after abdominal operations. In one patient, the remifentanil loading dose caused respiratory depression requiring intervention with O2 and ventilatory support by mask for 5 min. This suggests that special attention must be given to respiration during follow-up of patients when initiating remifentanil PCA.

Acute tolerance to opioids and its prevention remain controversial [16]. According to several animal studies, acute tolerance to the analgesic effect of opioids can develop [17-19]. However, the occurrence of this phenomenon during remifentanil treatment in human beings is somewhat controversial in clinical settings. Although there are some studies to support the claim that acute tolerance during remifentanil infusion develops [20,21], it does not necessarily means that this phenomenon can be seen in every patient in different clinical situations [22,23]. A recent randomized, placebo-controlled double-blinded crossover study in volunteers failed to demonstrate acute tolerance to remifentanil during a 3 h infusion of remifentanil at 0.08 μg kg−1 min−1 [24]. Our results are in agreement with the findings of Schraag and colleagues [22] who found no evidence of tolerance to remifentanil in postoperative patient-maintained, target-controlled infusions of the opioid during 6 h after surgery.

A randomized study conducted by Thurlow and colleagues [25] examined the effectiveness of i.v. remifentanil via PCA (20 μg bolus, 3 min lockout, and no background infusion) compared with i.m. meperidine 100 mg with an antiemetic drug for pain relief in 36 labouring females. The remifentanil PCA provided better pain relief to mothers in labour than the meperidine judged by VAS pain scores and overall assessment by mothers and attending midwives. Although there was no significant difference in the recorded minimum ventilatory rate and minimum SaO2, they concluded that it was a potent respiratory depressant and recommended adequate continuous monitoring.

Use of a remifentanil PCA for a patient with multiple rib fractures has been reported [26]. Remifentanil was administered from a PCA pump with a bolus dose of 25 μg (approximately 0.5 μg kg−1) using a 5-min lockout time. On day 6 a continuous infusion of 50 μg h−1 (approximately 1 μg kg−1 h−1) was added as a background infusion to the PCA. It was suggested that the use of remifentanil as a PCA be considered in patients where rapid control of analgesia is required, for example, for chest physiotherapy, and the accumulative sedative and respiratory depressant effects of longer-acting opiates are undesirable.

Volmanen and colleagues [27] conducted a study to evaluate the minimum effective analgesic dose of remifentanil and to assess the safety of short-term remifentanil when it is administered for pain relief during the first stage of labour. There was wide individual variation in the dose required for effective labour analgesia. The median effective PCA bolus was 0.4 μg kg−1 (range 0.2-0.8 μg kg−1) and consumption was 0.066 μg kg−1 min−1 (0.027-0.207 μg kg−1 min−1).

In this study, we employed a continuous background infusion of the opioid, since the rapid offset of remifentanil's action might hamper its successful use. The continuous background infusion of morphine during PCA has been studied extensively and shown to increase morphine consumption, sedation and respiratory depression without improving pain relief or patient satisfaction [28-30]. On the other hand, following abdominal surgery, a continuous infusion of morphine at 1 mg h−1 in addition to the PCA administration (morphine 1 mg bolus dose and 5 min lockout period) improved analgesia during the first 24 h but was associated with a greater incidence of complications than PCA alone [31]. In the light of all this, although the advisability of combining a continuous infusion with PCA is uncertain for morphine, it may provide an advantage for remifentanil.

In summary, the results suggest that with careful monitoring remifentanil PCA can be added to our current armamentarium for postoperative pain management after abdominal surgery when remifentanil becomes as cost effective as morphine in the future. Nevertheless, more controlled studies concerning its use with different dose settings are needed to establish solid criteria for its clinical applications.


1. Walder B, Schafer M, Henzi I, Tramer MR. Efficacy and safety of patient-controlled opioid analgesia for acute postoperative pain: a quantitative systematic review. Acta Anaesthesiol Scand 2001; 45: 795-804.
2. Unlugenc H, Ozalevli M, Guler T, Isik G. Postoperative pain management with intravenous patient-controlled morphine: comparison of the effect of adding magnesium or ketamine. Eur J Anaesthesiol 2003; 20: 416-421.
3. Lehmann KA. New developments in patient-controlled postoperative analgesia. Ann Med 1995; 27: 271-282.
4. Owen H, Plummer J. Patient-controlled analgesia: current concepts in acute pain management. CNS Drugs 1997; 8: 203-218.
5. Etches RC. Patient-controlled analgesia. Surg Clin North Am 1999; 79: 297-312.
6. Feldman PL, James MK, Brackeen MF, et al. Design, synthesis and pharmacological evaluation of ultrashort- to long-acting opioid analgesics. J Med Chem 1991; 34: 2202-2208.
7. James MK. Remifentanil and anesthesia for the future. Exp Opin Invest Drugs 1994; 3: 331-340.
8. Glass PSA, Hardman D, Kamiyama Y, et al. Preliminary pharmacokinetics and pharmacodynamics of an ultra-short-acting opioid: remifentanil (G187084B). Anesth Analg 1993; 77: 1031-1040.
9. Kapila A, Glass PSA, Jacobs JR, et al. Measured context-sensitive half-times of remifentanil and alfentanil. Anesthesiology 1995; 83: 968-975.
10. Davis PJ, Stiller RL, Wilson AS, et al. In vitro remifentanil metabolism: the effects of whole blood constituents and plasma butyrylcholinesterase. Anesth Analg 2002; 95: 1305-1307.
11. Ramsay MA, Savage TM, Simpson BR, Goodwin R. Controlled sedation with alphaxalone-alphadolone. Br Med J 1974; 2: 656-659.
12. Wrench IJ, Cavill G, Ward JEH, Crossley AWA. Comparison between alfentanil, pethidine and placebo in the treatment of postanesthetic shivering. Br J Anaesth 1997; 79: 541-542.
13. Hunt R, Fazekas B, Thorne D, Brooksbank M. A comparison of subcutaneous morphine and fentanyl in hospice cancer patients. J Pain Symptom Manage 1999; 18: 111-119.
14. Yarmush J, D'Angelo R, Kirkhart B, et al. A comparison of remifentanil and morphine sulfate for acute postoperative analgesia after total intravenous anesthesia with remifentanil and propofol. Anesthesiology 1997; 87: 235-243.
15. Ramsay MA, Macaluso A, Tillmann Hein HA, Cancemi E. Use of remifentanil in patients breathing spontaneously during monitored anesthesia care and in the management of acute postoperative care. Anesthesiology 1998; 88: 1124-1126.
16. Servin FS. Remifentanil: an update. Curr Opin Anaesthesiol 2003; 16: 367-372.
17. Gardmark M, Ekblom M, Bouw R, Hammarlund-Udenaes M. Quantification of effect delay and acute tolerance development to morphine in the rat. J Pharmacol Exp Ther 1993; 267: 1061-1067.
18. Kissin I, Bright CA, Bradley Jr EL. The effect of ketamine on opioid-induced acute tolerance: can it explain reduction of opioids consumption with ketamine-opioid analgesic combinations? Anesth Analg 2000; 91: 1483-1488.
19. Hayashida M, Fukunaga A, Hanaoka K. Detection of acute tolerance to the analgesic and nonanalgesic effects of remifentanil infusion in a rabbit model. Anesth Analg 2003; 97: 1347-1352.
20. Guignard B, Bossard AE, Coste C, et al. Acute opioid tolerance: intraoperative remifentanil increases postoperative pain and morphine requirement. Anesthesiology 2000; 93: 409-417.
21. Vinik HR, Kissin I. Rapid development of tolerance to analgesia during remifentanil infusion in humans. Anesth Analg 1998; 86: 1307-1311.
22. Schraag S, Checketts MR, Kenny GN. Lack of rapid development of opioid tolerance during alfentanil and remifentanil infusions for postoperative pain. Anesth Analg 1999; 89: 753-757.
23. Cortinez LI, Brandes V, Munoz HR, Guerrero ME, Mur M. No clinical evidence of acute opioid tolerance after remifentanil-based anaesthesia. Br J Anaesth 2001; 87: 866-869.
24. Gustorff B, Nahlik G, Hoerauf KH, Kress HG. The absence of acute tolerance during remifentanil infusion in volunteers. Anesth Analg 2002; 94: 1223-1228.
25. Thurlow JA, Laxton CH, Dick A, et al. Remifentanil by patient-controlled analgesia compared with intramuscular meperidine for pain relief in labour. Br J Anaesth 2002; 88: 374-378.
26. Dill-Russell PC, Ng L, Ravalia A. Use of a remifentanil PCA for a patient with multiple rib fractures. Can J Anaesth 2002; 49: 757.
27. Volmanen P, Akural EI, Raudaskoski T, Alahuhta S. Remifentanil in obstetric analgesia: a dose-finding study. Anesth Analg 2002; 94: 913-917.
28. Baxter AD. Respiratory depression with patient-controlled analgesia (Editorial). Can J Anaesth 1994; 41: 87-90.
29. Parker RK, Holtmann B, White PF. Patient-controlled analgesia: does a concurrent opioid infusion improve pain management after surgery? JAMA 1991; 266: 1947-1952.
30. Russell AW, Owen H, Ilsley AH, et al. Background infusion with patient-controlled analgesia: effect on postoperative oxyhaemoglobin saturation and pain control. Anaesth Intens Care 1993; 21: 174-179.
31. Dawson PJ, Libreri FC, Jones DJ, et al. The efficacy of adding a continuous intravenous morphine infusion to patient-controlled analgesia (PCA) in abdominal surgery. Anaesth Intens Care 1995; 23: 453-458.

ANALGESIA; patient-controlled; ANALGESICS; remifentanil; ANALGESICS; OPIOID; morphine; PAIN; postoperative; SURGERY; abdominal

© 2005 European Society of Anaesthesiology