Abdominal surgery is associated with intense postoperative pain and large doses of opioids are needed to provide adequate analgesia . However, there is no reliable method of determining how much opioid a patient will require for effective analgesia or how much opioid will result in adverse effects. Theoretically co-administration of an opioid and an adjuvant reduces the amount of both opioid and adjuvant required for optimal pain relief, and thus may decrease the incidence of adverse effects [2–4]. Recent studies suggested a role for N-methyl-D-aspartate (NMDA) receptor antagonists, e.g. magnesium or ketamine, in the management of postoperative pain, as NMDA receptor antagonism inhibits induction and maintenance of central sensitization after nociceptive stimuli . This class of drugs has been recommended as primary or adjuvant agents in postoperative pain management [1,5]. However, few clinical studies have evaluated the benefit of adding NMDA antagonists to opioids [1,6,7]. For instance, Stubhaug and colleagues  showed that a 48 h continuous administration of low-dose ketamine, together with patientcontrolled analgesia (PCA) with morphine, prolonged the time to the first use of PCA morphine and reduced cumulative morphine consumption. Other studies have demonstrated a marked decrease in opioid consumption and/or pain intensity by systemic  or epidural co-administration [10,11] of ketamine and opioids. In contrast, some clinical studies have failed to demonstrate any improvement in postoperative analgesia with ketamine [12–14] or with magnesium .
The present clinical study was designed to determine whether the addition of magnesium or ketamine to morphine for intravenous (i.v.) PCA resulted in improved analgesic efficacy and lower pain scores compared with morphine PCA alone after major abdominal surgery.
Following Faculty Ethics Committee approval and informed patient consent, 90 ASA I-II patients, aged 19-60 yr, scheduled for elective major abdominal surgery with general anaesthesia, were enrolled into this study. Preoperatively, patients were instructed in the use of the PCA device (Abbott Pain Management Provider®, Class II, Type CF; Abbott, North Chicago, IL, USA) and the verbal rating scale (VRS) for pain assessment. Exclusion criteria included the inability to use the PCA device, longterm use of opioid medications and a history of chronic pain syndromes. All patients were premedicated with i.v. midazolam 0.1 mg kg−1 60 min before operation. Standard monitors (pulse oximetry, automated blood pressure cuff, lead II electrocardiogram) were used during surgery.
Anaesthesia induction was performed with thiopental (5 mg kg−1) and maintained with sevoflurane 1.5-2% in a mixture of 66% nitrous oxide and 34% oxygen. Neuromuscular relaxation was induced either by an i.v. bolus of vecuronium or cisatracurium (0.1-0.2 mg kg−1) and maintained by bolus administration (0.03 mg kg−1) at 30 min intervals. Following endotracheal intubation, patients were allocated randomly to receive one of three PCA modes. The study medications were prepared as morphine 0.4 mg mL−1 (Group M, n = 30), morphine 0.4 mg mL−1 + MgSO4 30 mg mL−1 (Group MM, n = 30) and morphine 0.4 mg mL−1 + ketamine 1 mg mL−1 (Group MK, n = 30). Whenever milligrams are henceforth reported, they refer to the morphine dose and not to the other component of the study solution.
First, a standardized loading dose (0.05 mg kg−1) was given to the patients when the VRS ≥ 2. Patients were then allowed to use bolus doses of their study solution (0.0125 mg kg−1 every 20 min without time limit) with the PCA device. All patients received prophylactic ondansetron 4 mg i.v. It was planned to exclude patients from the study in case the patient's clinical condition would have necessitated any change in the analgesic technique or resetting of the PCA pump.
Levels of discomfort, pain and sedation, cumulative morphine consumption, and haemodynamic variables (systolic arterial pressure, diastolic arterial pressure, heart rate, peripheral oxygen saturation) were recorded by an anaesthetist of the pain management team who was blinded to the assignment of the patients. This was done before the start of PCA and at 15, 30, 60, 120 min, 6, 12 and 24 h after the start. The level of discomfort was assessed using an 11 point numerical rating scale (0: no discomfort, 10: extreme discomfort). Pain intensity was assessed using an 11 point VRS (0: no pain, 10: the worst pain imaginable). The level of sedation was assessed using a five point scale (0: alert, 4: deep sleep) . The type and duration of surgery, and any adverse effects (sedation >2, nausea or vomiting, diplopia, or hallucinations) were recorded.
Continuous variables such as patient characteristic data (gender, age, weight), and the type and duration of surgery were analysed using one-way ANOVA. For each continuous variable, a normal distribution was checked. When the data were not normally distributed, an appropriate non-parametric test was chosen. Ordinal data (VAS, sedation, discomfort values) were analysed using the Kruskal-Wallis test. If the test proved significant, a U-test with Bonferroni correction was used to evaluate differences between groups. Time-dependent intragroup data were analysed by Friedman's test. The Wilcoxon signed rank sum test was used to evaluate the differences within groups compared with baseline. Ratios of complications were analysed with a Χ2-test. Results are means (±SD) and medians (minimum to maximum). Statistical analyses were performed using the statistical package SPSS® v.9.0 (SPSS, Cary, NC, USA).
There were three drop-outs, two because of inadequate data collection (one each in Groups M and MM), and one protocol violation in the Group M. There were no differences between the groups with respect to patient characteristic data, and the type and duration of surgery (Table 1). There were no important variations or differences in the haemodynamic data and the oxygen saturation.
There were no statistical differences in average pain scores before the start of the PCA. However, pain scores were significantly lower in both Groups MM and MK at 15, 30 and 60 min compared with Group M (P < 0.001) (Table 2). There were no statistical differences in average sedation scores before the start of the PCA, and at any time thereafter (P < 0.001) (Table 2). There were no differences in average discomfort scores before the start of PCA administration. However, the level of discomfort was significantly higher in Group M compared with Groups MM and MK at 15, 30 and 60 min (P < 0.001); discomfort scores were similar in Groups MM and MK (Table 2). Within groups, VRS, comfort and sedation scores decreased significantly and consistently up to 60 min (P < 0.001) (Figs 1–3).
Cumulative morphine consumption after 12 and 24 h was significantly higher in Group M alone compared with Groups MM and MK (Table 2). After 24 h, cumulative morphine consumption was 8% less in Group MM and 4% less in Group MK compared with Group M alone. There was no significant difference between Groups MM and MK.
Nine patients in the Group M (30%), six in Group MM (20%) and five in Group MK (16.6%) complained of nausea for 24 h after surgery despite the prophylactic i.v. ondansetron (n.s.). Three patients in Group M (3.3%), two in Group MM (2.2%) and two in Group MK (2.2%) complained of urinary retention (n.s.). One patient in Group M had shivering that resolved spontaneously.
Intravenous PCA with morphine alone, morphine + magnesium, and morphine + ketamine were well accepted by patients. The addition- of ketamine or magnesium to morphine allowed a significant reduction in PCA morphine consumption and resulted in an improved analgesic efficacy. However, this effect was small. Possible mechanisms explaining the improved analgesic efficacy include a synergistic or additive interaction between morphine and the NMDA antagonists ketamine and magnesium. An interaction is supported by the lower VRS and discomfort scores in Groups MM and MK with similar morphine consumption.
The NMDA-receptor channel complex contains binding sites for non-competitive antagonists, e.g. ketamine and magnesium. As the activation of C-fibres leads to neuronal excitation, the use of ketamine or magnesium as adjuvants for acute pain treatment seems a logical choice . Magnesium blocks the NMDA channels in a voltage-dependent way. It also produces a dramatic reduction of the NMDA-induced currents . Conversely, in the absence of extracellular magnesium, the effect of NMDA agonists is enhanced . Major surgery may be followed by a significant decrease in serum magnesium concentrations . For this reason, the prevention of hypomagnesemia after surgery may be of importance in the reduction of pain intensity. The results of this study suggest that magnesium may indeed exert a specific antinociceptive effect via blockade of the NMDA-receptor complex or a non-specific effect via prevention of hypomagnesemia. In a double-blind study, patients receiving a preoperative bolus and postoperative infusion of magnesium sulphate had lower morphine requirements, less discomfort and less subjective sleep disturbance than control patients during the first 48 h postoperation . Ketamine has analgesic properties  and decreases postoperative pain, opioid requirements [6,21], wound hyperalgesia  and experimental ischaemic pain. However, adverse psychological effects may limit its clinical use [21,23]. In a study comparing a combination of morphine and ketamine via PCA with morphine PCA alone, postoperative pain scores were reduced in patients who received ketamine . However, Reeves and colleagues reported that small-dose ketamine combined with PCA morphine provided no benefit to patients undergoing major abdominal surgery .
It has been claimed that the effect of NMDA-blocking agents may only be apparent after the receptor-operated ion channel has been opened by nociceptive stimulation [15,24]. This was the reason why we started the PCA administration when the VRS ≥ 2 in all groups. It has been reported that magnesium and ketamine inhibit the NMDA system differently . Evidence suggests that both hyperalgesia after tissue injury and the development of opiate tolerance involve activation of the NMDA receptor and subsequent biochemical processes resulting in central sensitization . Sharing of NMDA receptor activation by both processes suggests that ketamine may substantially enhance opioid-induced antinociception .
In this study, the VRS and discomfort scores were lower in Groups MM and MK, suggesting that these patients experienced less pain or that they needed less study solution to achieve effective analgesia within the first 24 h compared with those who received morphine alone. The high discomfort scores in Group M at 15, 30 and 60 min may be related to the relative lack of analgesic efficacy of the PCA morphine regimen. These results suggest that both magnesium and ketamine regimens have an effect on comfort and analgesic efficacy during postoperative pain management.
Theoretically, it may be expected that the sedative effect of morphine is increased with the addition of ketamine or magnesium, even when small amounts are used. In our study, the doses of magnesium and ketamine were low enough to avoid possible sedative effects.
PCA morphine induced an incidence rate of nausea of 22% when used for the treatment of postoperative pain [26,27]. In this study, the addition of ketamine or magnesium did not significantly alter the incidence rate of nausea and vomiting, although the incidence rate of nausea was lower with ketamine (17 versus 30%). There are reports suggesting that the concomitant use of ketamine or magnesium with morphine may decrease the incidence rate of nausea and vomiting [6,28]. This is most likely an indirect antiemetic effect via a decreased consumption of morphine in patients receiving ketamine or magnesium.
It has been reported that small-dose ketamine has risks of producing vivid dreams and hallucinations . Edwards and colleagues used a dose-finding approach to study the combination of morphine and ketamine for postoperative pain in elderly patients and they found that vivid dreaming was a problem at high ketamine doses (7.8 μg kg−1 min−1) . In that study, the dose of ketamine (1.25-2.5 μg kg−1 min−1) used was low enough to avoid these side-effects.
In conclusion, in the immediate postoperative period, adding magnesium or ketamine to morphine for i.v. PCA led to a significantly lower consumption of morphine. These differences are unlikely to be of any clinical relevance. Further research is needed to support or refute this additive or synergistic effect.
1. Reeves M, Lindholm DE, Myles PS, Fletcher H, Hunt JO. Adding ketamine to morphine for patient-controlled analgesia after major abdominal surgery: a double-blinded, randomised controlled trial. Anesth Analg
2. McCarthy RJ, Kroin JS, Tuman. KJ, Penn RD, Ivankovich AD. Antinociceptive potentiation and attenuation of tolerance by intrathecal co-infusion of magnesium sulphate and morphine in rats. Anesth Analg
3. Kroin JS, McCarthy RJ, Roenn NV, Schwab B, Tuman KJ, Ivankovich AD. Magnesium sulphate potantiates morphine antinociception at the spinal level. Anesth Analg
4. Wilder-Smith CH, Knopfli R, Wilder-Smith OH. Perioperative magnesium infusion and postoperative pain. Acta Anaesthesiol Scand
5. Woolf CJ, Thompson SWN. The induction and maintenance of central sensitisation is dependent on N
-methyl-D-aspartic acid receptor activation: implications for the treatment of post-injury pain and hypersensitivity states. Pain
6. Adriaenssens G, Vermeyen KM, Hoffmann VL, Mertens E, Adriaensen HF. Postoperative analgesia with i.v. patientcontrolled morphine: effect of adding ketamine. Br J Anaesth
7. Ko SH, Lim HR, Kim DC, Han YJ, Choe H, Song HS. Magnesium sulphate does not reduce postoperative analgesic requirement. Anesthesiology
8. Stubhaug A, Breivik H, Eide PK, Kreunen M, Foss A. Mapping of punctuate hyperalgesia around a surgical incision demonstrates that ketamine is a powerful suppresser of central sensitisation to pain following surgery. Acta Anaesthesiol Scand
9. Javery KB, Ussery TW, Steger HG, Colclough GW. Comparison of morphine and morphine with ketamine for postoperative analgesia. Can J Anaesth
10. Chia YY, Liu K, Liu YC, Chang HC, Wong CS. Adding ketamine in a multimodal patient-controlled epidural regimen reduces postoperative pain and analgesic concentration. Anesth Analg
11. Wong CS, Liaw WJ, Tung CS, Su YF, Ho ST. Ketamine potentiates analgesic effect of morphine in postoperative epidural pain control. Reg Anesth
12. Dahl V, Ernoe PE, Steen T, Raeder JC, White PF. Does ketamine have preemptive effects in women undergoing abdominal hysterectomy procedures? Anesth Analg
13. Edwards ND, Fletcher A, Cole JR, Peacock JE. Combined infusions of morphine and ketamine for postoperative pain in elderly patients. Anaesthesia
14. Ilkjaer S, Nikolajsen L, Hansen TM, et al
. Effect of i.v. ketamine in combination with epidural bupivacaine or epidural morphine on postoperative pain and wound tenderness after renal surgery. Br J Anaesth
1998; 81: 707-712.
15. Dickenson AH. NMDA receptor antagonists: interaction with opioids. Acta Anaesthesiol Scand
16. Ascher P, Nowak L. Electrophysiological studies of NMDA receptors. Trends Neurosci
17. Garthwaite G, Garthwaite J. Receptor-linked ionic channels mediate N
-methyl-D-aspartate neurotoxicity in rat cerebellar slices. Neurosci Lett
18. Sanchez-Capuchino A, McConachie I. Peri-operative effect of major gastrointestinal surgery on serum magnesium. Anaesthesia
19. Tramèr MR, Schneider J, Marti RA, Rifat K. Role of magnesium sulfate in postoperative analgesia. Anesthesiology
20. Owen H, Reekie RM, Clements JA, Watson R, Nimmo WS. Analgesia from morphine and ketamine. A comparison of infusions of morphine and ketamine for postoperative analgesia. Anaesthesia
21. Edwards ND, Fletcher A, Cole JR, Peacock JE. Combined infusions of morphine and ketamine for postoperative pain in elderly patients Anaesthesia
22. Owen H, Plummer JL, Armstrong I, Mather LE, Cousins MJ. Variables of patient-controlled analgesia: 1. Bolus size. Anaesthesia
23. Mao J, Price DD, Mayer DJ. Mechanisms of hyperalgesia and morphine tolerance a current view of their possible interactions. Pain
24. Arendt-Nielsen L, Petersen-Felix S, Fisher M, Bjerring P, Zbinden AM. The effect of N
-methyl-D-aspartate antagonist (ketamine) on single and repeated nociceptive stimuli: a placebo controlled experimental human study. Anesth Analg
25. Wilder-Smith O, Arendt-Nielsen L, Gäumann D, Tassonyi E, Rifat KR. Sensory changes and pain after abdominal hysterectomy: a comparison of anesthetic supplementation with fentanyl versus magnesium or ketamine. Anesth Analg
26. Vickers MD, O'Flaherty D, Szekely SM, Read M, Yoshizumi J. Tramadol: pain relief by an opioid without depression of respiration. Anaesthesia
27. Vickers MD, Paravicini D. Comparison of tramadol with morphine for postoperative pain following abdominal surgery. Eur J Anaesthesiol
28. Watcha MF, White PF. Postoperative nausea and vomiting. Its etiology, treatment and prevention. Anesthesiology