The administration of morphine for postoperative analgesia after major surgery using IV patient-controlled analgesia (PCA) is a common practice. Self-administration of the opioid allows patients to satisfy their individual needs. However, opioid side effects, such as nausea and heavy sedation, are frequent.1 Respiratory depression, although rare, is a concern. Development of tolerance, i.e., resistance to the analgesic effect, might occur even at a very early stage of opioid therapy.2
Several clinical trials have shown that N-methyl-d-aspartate (NMDA) receptor antagonists have a role in modulating both acute and chronic somatic and visceral pain. When the NMDA receptors in the central nervous system are stimulated by afferent nociceptive input, they activate a neuronal sensitization process that enhances pain perception.3 NMDA-receptor activation not only increases the cells’ response to pain stimuli, it may also decrease neuronal sensitivity to opioid receptor agonists.4 In addition to preventing central sensitization, the co-administration of NMDA-receptor antagonists with an opioid may prevent tolerance to opioid analgesia.5
Randomized controlled studies have found that the addition of the NMDA-receptor antagonist ketamine to morphine decreases pain scores and side effects.6,7 However, subsequent randomized controlled studies showed no benefit to adding ketamine to morphine.8,9 Important limitations of the aforementioned studies are the low number of patients analyzed, the use of morphine-ketamine combinations that may not be optimal, and that not all the relevant outcomes have been analyzed. Meta-analyses10,11 did not clarify the role of ketamine as a component of postoperative analgesia. The trials were characterized by limited size, groups rarely exceeding 30 patients, wide variability in clinical settings, ketamine regimens, and outcome measures. Outcomes that might be regarded as clinically relevant (i.e., opioid-related adverse effects) were inconsistently reported. In the meta-analysis by Elia and Tramer,10 they reported that only five trials, including 284 adult patients, have studied the combination of morphine and ketamine for IV PCA. In their article, Subramaniam et al.11 added only one study to those reviewed by Elia and Tramer. As a result, data pooling was impossible for the majority of regimes or was deemed inadequate to address the issue of incidence of rare severe complications.10
The aim of this study was to compare morphine alone with a combination of morphine-ketamine for PCA. The combination in our study was found to be optimal in a previous study12 and proved safe when administered at the ward.13 A brief outline of the design and the results of our optimization study12 are given in the Methods. We analyzed a larger sample size than in our previous studies to detect possible differences in all clinically relevant outcomes.
After obtaining approval from the ethics committee of the University of Bern, Switzerland, patients undergoing major elective orthopedic surgery (Table 1) were studied. Written informed consent was obtained from all patients. Exclusion criteria were any contraindication to ketamine or morphine, age <18 yr, intake of psychotropic drugs, daily intake of opioids for a period >1 wk, and lack of patient’s cooperation. The patients, nurses who cared for patients, anesthesiologists who performed the anesthesia, and investigators who gathered the data were not aware of the PCA drug used. The drugs were simply labeled study drugs. Randomization was performed by drawing lots immediately before administering the solution.
Either general or regional anesthesia was performed, as decided by the anesthesiologist in charge of the patient.
Patients were premedicated orally with midazolam 7.5 mg, 20–30 min before anesthesia. They were monitored with electrocardiogram, noninvasive arterial blood pressure (one measurement every 5 min), and oxygen saturation using pulse oximetry.
General anesthesia was induced with IV fentanyl 0.15–0.2 mg; thiopental 5–7 mg/kg or propofol 2–2.5 mg/kg or etomidate 0.2–0.3 mg/kg; vecuronium 0.1 mg/kg or atracurium 0.5 mg/kg. After tracheal intubation, a mixture of oxygen (30% inspired concentration) with nitrous oxide was administered, supplemented by either isoflurane (0.3–0.5 vol% end-tidal concentration) or propofol (2–4 μg/mL target-controlled infusion). If there were signs of inadequate analgesia, IV boluses of 0.05–0.2 mg fentanyl and 1.0–2.0 mg vecuronium or atracurium 10–20 mg were administered at the discretion of the attending anesthesiologist. At the end of surgery, residual neuromuscular blockade was reversed with 2.5 mg neostigmine and 0.5 mg glycopyrrolate.
Regional anesthesia was performed as a single-shot injection of a maximum 50 mL of mepivacaine 1% with sodium bicarbonate or ropivacaine 0.75% for peripheral nerve blockade and 12.5–17.5 mg bupivacaine 0.5% for spinal anesthesia.
For general anesthesia, the trachea was extubated as soon as patients opened their eyes to verbal command and had sufficient respiration. If extubation was not performed within 1 h after the end of the operation, the patient was excluded from the study. For regional anesthesia, patients who did not require the first PCA dose within the first 2 h after the end of the operation were excluded from the study.
Rationale for Choosing the Studied Dose of Ketamine
The combination of morphine with ketamine used in this study is based on the possible optimal combination indicated in our previous study.12 In that study, a novel method was used to identify the optimal combination of morphine with ketamine, using a stepwise analysis of several different combinations.
Drug combinations are generally investigated by comparing two or more groups, each receiving a different combination. However, this approach is challenged by a serious problem: the number of possible combinations. If we combine 2 different drugs and analyze 2 doses for each drug, there are 22 = 4 different combinations. However, if the therapeutic range of the drugs under investigation is wide, we might want to analyze more combinations, e.g., 5. In this case, we had to analyze 52 = 25 different combinations. If we want to add an additional variable, e.g., another drug or the time interval between the doses, the number of possible combinations increases to 53 = 125. Therefore, only a small portion of all possible combinations is investigated in a randomized controlled trial. Such a trial allows conclusions pertaining to the combinations analyzed, and the optimal combination may not be tested.
Direct Search Optimization Method
The optimum is searched stepwise. Initially, few combinations are tested. On the basis of the results obtained, new combinations are identified stepwise and investigated, until the optimal one is reached. Basically, the information obtained by the analysis of the combinations at each step is used to move away from the “bad” combinations toward the “good” ones (the optimum). It is then not necessary to explore all possible combinations.
We implemented the method developed by Berenbaum14 to optimize a postoperative epidural regimen15,16 and IV PCA with morphine and ketamine.17
One hundred and two patients undergoing lumbar spine or hip surgery participated in our optimization study.12 Initially, eight combinations of morphine, ketamine, and lockout interval (i.e., minimal allowed time between two consecutive PCA boluses) were empirically chosen and investigated. To determine subsequent combinations, an optimization model was applied until three consecutive steps showed no decrease in pain score. We analyzed 12 combinations with an allowed morphine and ketamine range in PCA solution of 0–2 mg/mL and a lockout interval range of 5–12 min. During the optimization procedure, a reduction in mean pain scores with a low incidence of side effects was observed. The procedure converged to a morphine: ketamine ratio of 1:1 and a lockout interval of 8 min. This combination was used in the present study.
The end of surgery was considered the beginning of the postoperative study period, which included the entire hospital stay. Patients were instructed on the use of the PCA unit both on the day before surgery and after end of the operation. No sedation was administered during the study period. All patients received 2 g propacetamol IV every 6 h as basal analgesia.
Patients with ASA physical status more than two were kept in the recovery room until the morning after the operation. During this time, oxygen saturation using pulse oximetry was continuously measured. Patients were moved to the ward when cardiocirculatory and respiratory function were stable. Oxygen 2–4 L/min via nasal probe was administered to maintain an oxygen saturation of more than 93%.
Patients were randomly allocated to receive PCA consisting of either morphine 1.5 mg (Group M) or morphine with ketamine 1.5 mg of each (Group MK). In both groups, the PCA pump was programmed to deliver a maximum of 6 boluses per hour, with a lockout time (i.e., the minimum time allowed between two boluses) of 8 min.
The verbal rating score was recorded every 2 h during the first 6 h and every 4 h thereafter by asking patients to rate pain at rest and during mobilization as follows: 0 = no pain, 1 = mild, 2 = moderate, 3 = strong, and 4 = severe pain. Mobilization was defined as passively turning patients on their sides for nursing procedures. Adequate analgesia was defined as a score of 0 at rest and ≤2 during mobilization. Patients were instructed to press the PCA button when they experienced pain of any intensity at rest, or moderate, strong or severe pain during mobilization. If adequate analgesia was not obtained after six consecutive boluses, the PCA bolus was increased by 0.5 mg every hour, to a maximum of 2.5 mg. If adequate analgesia was not achieved, ketorolac 30 mg IV every 8 h was administered.
Sedation was recorded according to the following score: 0 = alert; 1 = drowsy; 2 = asleep, easily arousable to verbal commands, does not fall asleep during or immediately after conversation; 3 = asleep, opens the eyes to verbal command, falls asleep during or immediately after conversation; 4 = does not open eyes to verbal command. A maximum score of 3 during the first 12 postoperative hours or 2 during the subsequent observation period was acceptable. In the presence of higher scores, the PCA bolus was reduced by 0.5 mg every hour until the patient recovered from sedation to acceptable levels.
If a respiratory rate of <8 per minute for a period longer than 10 min was observed, the PCA pump was stopped until a respiratory rate of 8 per minute was reached. The PCA regimen was then restarted using a PCA bolus of 0.5 mg less than the previous one.
In the presence of nausea, with or without vomiting, ondansetron 4 mg IV was given and repeated if nausea did not disappear. If nausea was not relieved after the administration of the second dose, the PCA bolus was reduced by 0.5 mg every hour, until there was no nausea.
Pruritus was treated only if severe. The skin was examined to eliminate other causes of pruritus. Clemastinum 2 mg was administered. If this drug was not effective, the PCA bolus was reduced by 0.5 mg every hour, until pruritus was relieved.
Dreams and hallucinations were defined as any sensation that was not caused by an external event and were categorized as pleasant or unpleasant. In the presence of unpleasant dreams or any hallucinations (even pleasant), the PCA bolus was reduced by 0.5 mg every hour, until these symptoms were alleviated.
Unsatisfactory treatment was defined as the occurrence of one of the following situations: 1) inadequate analgesia: pain score >0 at rest and >2 during mobilization after 2.5 mg PCA boluses, repeated six times in 1 h; 2) side effect: (a) is not relieved despite reduction in PCA bolus, (b) is alleviated after reduction in PCA bolus, but analgesia is inadequate.
If the above protocol to deal with pain and side effects was not successful, individual pain treatment was planned. Modifications in the PCA drug combination, in the lockout interval, addition of other analgesics or change from PCA to another form of pain therapy were the available options.
The PCA was discontinued when patients required an average of <1 PCA bolus per hour during the last 12 h.
As the main outcome, the number of patients with unsatisfactory treatment (see above for definition) was calculated.
Data concerning the patient were gender, age, weight, ASA class and type of surgery.
Data concerning the intraoperative phase were type of anesthesia (general, regional or combined) amount of morphine and fentanyl administered and duration of operation.
In the postoperative phase, during the time in which PCA was used, the following data were assessed for each patient every 2 h during the first 6 h and every 4 h thereafter: pain intensity using the verbal rating score, sedation score, systolic blood pressure, heart rate, presence of dreams and hallucinations, respiratory rate, presence of nausea, vomiting, and pruritus. In addition, total consumption of PCA drugs and duration of the PCA use were recorded. Direct medical costs were calculated from the total amount of additional ketorolac or metamizole, type and dose of drugs to treat the side effects and acute pain service team’s additional time to manage inadequate analgesia and adverse effects. To assess the number of patients with chronic postoperative pain, patients were mailed a form 3 and 6 mo after the operation. Patients were asked if they still had pain at the same location as before the operation and, if so, to rate the intensity on a visual analog scale. Furthermore, patients were asked whether their pain was less, equal or worse than the pain before surgery.
The sample size was calculated considering the proportion of patients with unsatisfactory treatment. In a previous investigation,12 unsatisfactory treatment was observed in 5 of 48 patients receiving combinations of morphine-ketamine categorized as optimal. In those combinations, the PCA bolus dose was close to the one selected in the present study. We considered a difference of 0.125 in the proportion of unsatisfactory treatment between the groups as clinically relevant, i.e., 6/48 = 0.125 in the Group MK and 0.25 in Group M. We analyzed the data using the χ2-test. Setting [alpha] = 0.05 and [beta] = 0.80, a statistically significant difference would be detected by analyzing 168 patients per group.
Backward stepwise regression was used to identify the predictors of unsatisfactory treatment, with unsatisfactory treatment as dependent variable and the following independent variables: drugs used, type of surgery, type of anesthesia, intraoperative administered fentanyl, ASA class, maximum bolus administrated, pain score, duration of PCA therapy, and consumption of PCA drugs.
All the other parameters, numerical and categorical data were compared by the Student’s t-test and the χ2 test, respectively.
Since analyzing so many secondary outcome parameters may provide a statistically significant difference as the result of pure chance, we deliberately used descriptive statistics only and decided not to use Bonferroni correction of the Student’s t-test.
Of the 401 patients enrolled, 49 were not included in the analyses for the following reasons: intraoperative protocol violation (23 patients), postoperative protocol violation,14 postoperative use of indometacin for ectopic ossification prophylaxis,11 surgical revision.1 The study was therefore completed in 352 patients. The two groups were comparable (Table 2).
There were no differences between the groups with respect to the primary end-point (unsatisfactory treatment, Table 3). The incidence of respiratory depression was more frequent in Group MK, i.e., 10.8% vs 5.1% in Group M (P = 0.075) as well as the need to administer additional analgesic and ondansetron (Table 3). No differences were found in other secondary end points: adverse effects, pain scores, consumption of administered drugs (Fig. 1), duration of PCA therapy, additional visits from the acute pain team (Table 3) or incidence and intensity of chronic pain (Table 4). Only 25.9% and 10.2% of all patients returned our chronic pain questionnaire 3 and 6 mo after the surgery, respectively.
Results of backward stepwise regression show that unsatisfactory treatment can be predicted from pain score at rest, duration of PCA therapy, and the consumption of PCA drugs. PCA regimen used, type of surgery, type or duration of anesthesia, ASA class, amount of drug administered per bolus, and intraoperative consumption of fentanyl were not significant predictors and were progressively removed from the model (Table 5).
The available literature on a morphine-ketamine combination for postoperative IV PCA is conflicting (Table 6). Studies on a morphine-ketamine combination after major abdominal surgery investigated a small number of patients and no study has been performed in a large number of patients who underwent orthopedic surgery.18 In this double-blind, randomized trial with a large number of patients, we rejected the hypothesis that small-dose ketamine is a useful adjunct to morphine PCA after major elective orthopedic surgery.
The combination of morphine with ketamine used in this study was based on the possible optimal combination indicated in our previous study in which patients undergoing major hip and spine surgery were investigated.12 We therefore assumed that the combination found could be generalized for patients undergoing major orthopedic surgery.
There was no significant difference between groups regarding primary or secondary outcome variables. Pain scores recorded were low (Table 3). Higher demand for additional analgesic in the Group MK was not statistically significant and might have been the result of pure chance.
Edwards et al.19 found no beneficial effect of ketamine on postoperative lung function, even though the bronchodilator effect of ketamine is well known. Interestingly, in our study, respiratory depression was more frequent, although not statistically significant, in the Group MK than in Group M alone.
In general, the incidence of side effects in our study was more frequent than the incidence reported in the literature (Table 3).10,11,18 It is possible that the incidence of side effects was under-estimated in previous studies.
The average ketamine consumption in the present study was 1.86 mg/h, which is comparable with the results of previous studies.10 In the study conducted by Javery et al.7 the average ketamine consumption was 1.2 mg/h and the addition of ketamine produced a remarkable effect after elective microdiscectomy. In the study by Reeves et al.,8 the average ketamine consumption in the first 24 h after major abdominal surgery was 3.2 mg/h, yet they did not find any measurable beneficial effect. The same average ketamine consumption was observed by Murdoch et al.20 in a study on a small patient group that underwent total abdominal hysterectomy. This study also failed to show any benefit of adding ketamine to morphine PCA. In a study of 30 patients, Adrieanssens et al.6 showed that 10 mg/h of ketamine had a morphine-sparing effect with reduction of nausea after abdominal surgery. Besides Javery et al.7 there is no study using morphine and ketamine PCA for treatment of musculoskeletal pain.
The smallest ketamine plasma concentration to counteract hyperalgesia while producing minimal side effects was shown to be 60 μg/mL.21 This concentration was achieved by giving an initial bolus dose of ketamine 0.5 mg/kg, followed by a continuous infusion of 2 μg · kg−1 · min−1.22,23 In comparison, the average ketamine consumption in our study was more than six times lower, which might explain the negative result. In addition, continuous infusion of small-dose ketamine during PCA with morphine may be better than a PCA approach alone, because of a stable NMDA receptor block. On the other hand, it would reduce patients’ mobility, increase costs and may be less safe.24 Whether a continuous ketamine infusion is superior to PCA delivery is yet to be demonstrated by a large, randomized, controlled trial.
Because of the very low return rate of our questionnaire regarding chronic pain (only 10.2% of all patients returned a questionnaire 6 mo after the surgery), no conclusions can be made regarding the influence of ketamine on chronic pain, a more active manner of collecting long-term data is needed to assure quality data collection.
Further studies are needed in subgroups of patients to investigate if the addition of ketamine to an opioid may be considered as adjunctive therapy instead of routine use. Patients with morphine-resistant pain,25 opioid tolerance,26 or addiction27 may be possible target groups.
In conclusion, we failed to demonstrate any beneficial effect of routinely adding ketamine to morphine for PCA after major orthopedic surgery.
1. Tsui SL, Irwin MG, Wong CM, Fung SK, Hui TW, Ng KF, Chan WS, O’Reagan AM. An audit of the safety of an acute pain service. Anaesthesia 1997;52:1042–7
2. Vinik HR, Kissin I. Rapid development of tolerance to analgesia during remifentanil infusion in humans. Anesth Analg 1998;86: 1307–11
3. Dickenson AH, Sullivan AF. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of deep rat dorsal horn nociceptive neurones following C fibre stimulation. Neuropharmacology 1987;26:1235–8
4. Dickenson AH. NMDA receptor antagonists: interactions with opioids. Acta Anaesthesiol Scand 1997;41:112–5
5. Miyamoto H, Saito Y, Kirihara Y, Hara K, Sakura S, Kosaka Y. Spinal coadministration of ketamine reduces the development of tolerance to visceral as well as somatic antinociception during spinal morphine infusion. Anesth Analg 2000;90:136–41
6. Adriaenssens G, Vermeyen KM, Hoffmann VL, Mertens E, Adriaensen HF. Postoperative analgesia with i.v. patient-controlled morphine: effect of adding ketamine. Br J Anaesth 1999;83:393–6
7. Javery KB, Ussery TW, Steger HG, Colclough GW. Comparison of morphine and morphine with ketamine for postoperative analgesia. Can J Anaesth 1996;43:212–5
8. 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, randomized controlled trial. Anesth Analg 2001;93:116–20
9. Burstal R, Danjoux G, Hayes C, Lantry G. PCA ketamine and morphine after abdominal hysterectomy. Anaesth Intensive Care 2001;29:246–51
10. Elia N, Tramer MR. Ketamine and postoperative pain – a quantitative systematic review of randomised trials. Pain 2005;113:61–70
11. Subramaniam K, Subramaniam B, Steinbrook RA. Ketamine as adjuvant analgesic to opioids: a quantitative and qualitative systematic review. Anesth Analg 2004;99:482–95
12. Sveticic G, Gentilini A, Eichenberger U, Luginbuhl M, Curatolo M. Combinations of morphine with ketamine for patient-controlled analgesia: a new optimization method. Anesthesiology 2003;98:1195–205
13. Sveticic G, Eichenberger U, Curatolo M. Safety of mixture of morphine with ketamine for postoperative patient-controlled analgesia: an audit with 1026 patients. Acta Anaesthesiol Scand 2005;49:870–5
14. Berenbaum MC. Direct search methods in the optimisation of cancer chemotherapy regimens. Br J Cancer 1990;61:101–9
15. Curatolo M, Schnider T, Petersen-Felix S, Weiss S, Signer C, Scaramozzino P, Zbinden A. A direct search procedure to optimize combinations of epidural bupivacaine, fentanyl, and clonidine for postoperative analgesia. Anesthesiology 2000;92:325–37
16. Sveticic G, Gentilini A, Eichenberger U, Zanderigo E, Sartori V, Luginbuhl M, Curatolo M. Combinations of bupivacaine, fentanyl, and clonidine for lumbar epidural postoperative analgesia: a novel optimization procedure. Anesthesiology 2004;101:1381–93
17. Sveticic G, Gentilini A, Eichenberger U, Luginbuhl M, Curatolo M. Combinations of morphine with ketamine for patient-controlled analgesia: a new optimization method. Anesthesiology 2003;98:1195–205
18. Himmelseher S, Durieux ME. Ketamine for perioperative pain management. Anesthesiology 2005;102:211–20
19. Edwards ND, Fletcher A, Cole JR, Peacock JE. Combined infusions of morphine and ketamine for postoperative pain in elderly patients. Anaesthesia 1993;48:124–7
20. Murdoch CJ, Crooks BA, Miller CD. Effect of the addition of ketamine to morphine in patient-controlled analgesia. Anaesthesia 2002;57:484–8
21. Leung A, Wallace MS, Ridgeway B, Yaksh T. Concentration-effect relationship of intravenous alfentanil and ketamine on peripheral neurosensory thresholds, allodynia and hyperalgesia of neuropathic pain. Pain 2001;91:177–87
22. Stubhaug A, Breivik H, Eide PK, Kreunen M, Foss A. Mapping of punctuate hyperalgesia around a surgical incision demonstrates that ketamine is a powerful suppressor of central sensitization to pain following surgery. Acta Anaesthesiol Scand 1997;41:1124–32
23. Schmid RL, Sandler AN, Katz J. Use and efficacy of low-dose ketamine in the management of acute postoperative pain: a review of current techniques and outcomes. Pain 1999;82:111–25
24. Van Elstraete AC, Delanoue K, Fuzier V. Combination of morphine with ketamine for patient-controlled analgesia: is ketamine plasma concentration adequate? Anesthesiology 2004;100: 197–8
25. Weinbroum AA. A single small dose of postoperative ketamine provides rapid and sustained improvement in morphine analgesia in the presence of morphine-resistant pain. Anesth Analg 2003;96:789–95
26. de Leon-Casasola OA. Cellular mechanisms of opioid tolerance and the clinical approach to the opioid tolerant patient in the post-operative period. Best Pract Res Clin Anaesthesiol 2002;16:521–5
27. Haller G, Waeber JL, Infante NK, Clergue F. Ketamine combined with morphine for the management of pain in an opioid addict. Anesthesiology 2002;96:1265–6