Multimodal analgesia consists of concurrent administration of 2 or more analgesic drugs that act by different mechanisms and, ideally, at different levels of the nociceptive pathway. Multimodal analgesia often results in additive and sometimes synergistic effects.1–4 Epidural administration of analgesic agents is a common technique used in animals to alleviate pain, particularly in the pelvic limbs. Of the several combinations of opioids and local anesthetics that have been studied, the morphine-bupivacaine combination has gained popularity. The literature includes numerous studies5–13 in which the efficacy of epidurally administered opioids (primarily morphine and oxymorphone) and local anesthetics (primarily bupivacaine and lidocaine), alone or in combination, was investigated. The analgesia induced by the combinations was generally superior to that induced by each class of agents administered alone.5–13 This was particularly true in experimental evaluations involving rodent models6,8,12 and in situations involving preemptive epidural administration of the mixture.7,10,11,13,a With preemptive analgesia, analgesic drugs are administered before initiation of the nociceptive stimulation that leads to adverse CNS changes called central sensitization to pain. The fact that it is easier to prevent central sensitization before the painful stimulus is initiated (ie, to induce preemptive analgesia) than it is to treat the subsequent hyperalgesia with curative analgesia appears to be well established.14 To be most effective, preemptive analgesia must prevent noxious information from reaching the CNS. Indeed, it has been recognized that established pain can only be controlled, as opposed to eliminated.15 In that respect, preemptive analgesia is slightly different from preventive analgesia, in which the effects include both the surgical and initial postoperative periods. Preventive analgesia is evaluated by comparing the effects of analgesic drugs and a placebo when administered before initiation of the nociceptive stimulation. Past studies16–21 have yielded conflicting results regarding the analgesic efficacy of epidurally administered combinations involving opioids and local anesthetics. Particular experimental conditions, such as timing (pre- vs postoperative administration17–19,21,22), bupivacaine concentration,18,19 provision of additional intraoperative analgesics,17,18 method of assessing analgesia, and physical activity of the patients,16,23,24 may explain some of the discrepancies.
For postoperative curative analgesia after orthopedic surgery in dogs anesthetized with isoflurane, Hendrix et al9 reported the superiority of the epidurally administered morphine-bupivacaine combination over administration of morphine, bupivacaine, or isotonic saline (0.9% NaCl) solution alone. In that study,9 epidural injection was performed after surgery, shortly before the end of general anesthesia. The efficacy of morphine was not superior to that of placebo, a finding that seems to be in contradiction with present clinical impressions and results of other studies.7,13,25,26 Because the epidural injection was administered after surgery,9 it was not possible to evaluate its effect on isoflurane requirements. Interestingly, all plasma cortisol concentrations were within reference range, and there were no significant differences between groups in that study.9
To our knowledge, there are no comparative clinical studies in which evaluation of the intraand postoperative efficacy of epidurally administered saline solution, morphine alone, or morphine combined with bupivacaine was conducted in dogs under actual clinical conditions (ie, administered before surgery in dogs that also received a parenterally administered opioid premedication). Several earlier studies9,10,13,a yielded similar results but differed from the present study in various aspects of protocol. The objective of the present study was to quantify the intra- and postoperative analgesic effects of 3 preemptive epidural analgesic protocols when administered in addition to a premedication protocol typically used in dogs undergoing elective orthopedic surgery. With regard to evaluation of potential adverse effects, our primary interest was the degree of impact on cardiovascular function (such as hypotension) induced by the epidural protocols, particularly the combination that included bupivacaine.
Materials and Methods
Dogs—Thirty-six dogs admitted to the Centre Hospitalier Universitaire Vétérinaire to undergo femoral head and neck excision or cranial cruciate ligament repair by lateral suture stabilization (via the modified Flo technique) were randomly selected and enrolled from 1999 to 2004. Exclusion criteria were dermatosis at the epidural injection site, coagulopathy, treatment with analgesics < 24 hours before anesthesia, age < 8 months, traumatic conditions (because this could interfere with evaluation of signs of stress or pain), and conditions constituting a contraindication to the use of any drug included in the protocol (eg, a history of seizures would have contraindicated administration of acepromazine) or that necessitated any treatment that could interfere with evaluation of the dogs (eg, administration of phenobarbital or antidepressant drugs). The study was approved by the Institutional Animal Care and Use Committee of the Faculty of Veterinary Medicine, and all owners provided informed consent.
Experimental protocol—A standardized anesthetic protocol was used. The protocol included premedication with IM administered oxymorphone hydrochlorideb (0.1 mg/kg [0.045 mg/lb]) or hydromorphone hydrochloridec (0.2 mg/kg [0.09 mg/lb]), acepromazine maleated (0.05 mg/kg [0.023 mg/lb]), and glycopyrrolatee (0.01 mg/kg [0.0045 mg/lb]). Induction proceeded with IV administration of thiopental sodiumf (a 10 mg/kg [4.54 mg/lb] dose was drawn and administered to effect). Dogs were orotracheally intubated, and anesthesia was maintained with isofluraneg in oxygen delivered by an out-of-circuit precision vaporizer. Initial vaporizer settings of 2.0% to 2.5% were decreased in increments of 0.25% to 0.5% every 5 to 10 minutes as long as the depth of anesthesia was deemed satisfactory on the basis of cardiovascular variables and mandibular muscle tone, loss of palpebral reflex, and absence of response to surgical stimulation.
Ventilation was mechanically controlled in all dogs to maintain end-tidal CO2 partial pressures of 30 to 40 mm Hg. Lactated Ringer's solution was administered IV at a flow rate of 5 to 10 mL/kg/h (2.27 to 4.54 mL/lb/h). Cardiovascular support consisted of IV administered glycopyrrolate (5 mg/kg [2.27 mg/lb]) or dobutamineh (1 to 5 mg/kg/min [0.45 to 2.27 mg/lb/min]), depending on whether bradycardia or hypotension, respectively, was detected. Bradycardia was defined as a heart rate < 55 beats/min, and hypotension was defined as mean systemic arterial blood pressure < 60 mm Hg for longer than 5 consecutive minutes.
Three epidural treatment protocols were investigated. One group of dogs received a combination of morphine sulfatei (0.2 mg/kg) and bupivacaine hydrochloridej (1 mg/kg) diluted in isotonic salinek solution to a total volume of 0.21 mL/kg (0.095 mL/lb), with a maximum limit volume of 6 mL. The second group of dogs received morphine sulfate (0.2 mg/kg) diluted in isotonic saline solution to a total volume of 0.21 mL/kg, with a maximum limit volume of 6 mL. Control dogs received a placebo treatment of 0.21 mL/kg of isotonic saline solution, with a maximum limit volume of 6 mL.
The same investigator (JJKB), blinded to treatment groupings, performed all epidural injections and the pre-, intra-, and postoperative evaluations. The loss-of-resistance technique was used to verify the needle's placement in the epidural space; 3 to 4 mL of saline solution was placed in a low-resistance syringe,l and an epidural catheterm was inserted for confirmation only. The catheter was removed before the epidural injection. It has been suggested that this method is more reliable than use of the loss-of-resistance method alone or the hanging-drop technique.13 Dogs were excluded from the study if the epidural space was not successfully detected with the loss-of-resistance method or if an epidural catheter could not be inserted.
Preoperative Evaluation
Behavior—The dogs' general behavior and tolerance to restraint were preoperatively evaluated during venipuncture by use of a variation of a scoring method described elsewhere (Appendix 1).27,28 Before any manipulation, including venipuncture, dogs were assessed by means of simple visual observation (yielding an attitude score). Behavior was also evaluated when the blood sample was collected (yielding a cooperation score).
Plasma cortisol concentration—Plasma cortisol concentration was determined as a nonspecific indicator of stress and pain and was assayed from samples collected before surgery (T0) to determine a baseline value. Blood samples were drawn into 3-mL heparinized tubes before anesthesia (time varied according to availability of the dogs) and at various times thereafter. Samples were refrigerated for £ 2 hours before being frozen at −70°C for up to 11 months. Cortisol analyses were performed via radioimmunoassayn and chemiluminescenceo (the latter was used after the former technique became unavailable).
Intraoperative Evaluation
Rescue analgesia and end-tidal isoflurane concentration—When heart rate or mean arterial pressure increased by > 30% of the preoperative value, end-tidal isoflurane concentration was increased in increments of 0.25% to 0.5%. Isoflurane requirements were evaluated by measuring endtidal concentration with a gas analyzer.p If the hemodynamic change was sustained for > 5 minutes despite the increase in isoflurane or when signs obviously caused by surgical stimulation were observed (eg, a marked and sudden increase in heart rate, mean arterial pressure, or both; abdominal ventilation; or movement), rescue analgesia consisting of hydromorphone (0.1 to 0.2 mg/kg) or oxymorphone (0.05 to 0.1 mg/kg), IV, was administered.
Hemodynamics—Arterial blood pressure was measured invasively by use of a catheterq placed in a dorsal pedal artery after the dog entered the operating room. Catheters were placed before surgical draping and shortly after anesthetic induction and were connected to a transducerr and oscilloscope.s Dogs in which arterial catheterization was not completed were excluded from the study. Heart rate was derived from the ECG or pulse pressure curve if the former was not accurate.
Other—End-tidal CO2 partial pressure and respiratory rate were obtained from the gas analyzer display. Pulse oximetryt was used when available, and arterial blood gas analysisu was performed when necessary.
Postoperative Evaluation
Pain score—Postoperative evaluations were conducted every hour for the first 12 hours and again at 24 hours. A pain scoring system described elsewhere27,28 and modified with permission of the authors was used (Appendix 2). Data were collected in the following sequence: visual behavioral evaluation, respiratory rate, heart rate and mean arterial pressure, and the remainder of the components of behavioral evaluation that involved physical interaction. The item on the scoring system involving manipulation of the surgical wound was analyzed individually and was called the manipulation score to differentiate it from the total pain score. The manipulation score was used to differentiate signs of dysphoria from signs of true pain.
Rescue analgesia—Hydromorphone or oxymorphone was administered IM or IV at the same doses as were used during surgery in dogs with a pain score > 5 out of a possible 15. If a high total pain score was attributable to dysphoria (ie, higher than 5/15 but with a score of 0 for the item involving manipulation of the surgical wound), a tranquilizer (acepromazine at a dose of 0.05 mg/kg or diazepam at a dose of 0.2 mg/kg) was administered IV.
Plasma cortisol concentration—Blood samples were drawn into 3-mL heparinized tubes at T0 (which corresponded to the end of anesthesia, when the isoflurane vaporizer was turned off) and again at 2, 4, 6, 8, 12, and 24 hours. Samples were processed and analyzed as described.
Statistical analysis—Variables were evaluated, and pertinent statistics were summarized by use of commercially available softwarev (Appendix 3). Most of the continuous quantitative variables were analyzed by use of the general linear model. When a significant difference was detected with > 2 categories, the Tukey post hoc test was used to determine honest significant differences between pairs of categories. Some quantitative ordinal variables, such as scores, were also analyzed with the linear model when the possible number of values was sufficiently high. The exact χ2 test was used with binary variables (ie, administration or no administration of rescue analgesia) and for categoric or quantitative ordinal variables. Quantitative discontinuous variables were analyzed by use of negative binomial regression. Among treatment groups, survival analysis was used to compare delays before the time at which rescue analgesia was deemed necessary. Differences were considered significant at values of P < 0.05.
Results
Demographic data—Twenty-two female dogs and 14 males were randomly distributed among the treatment groups as follows: 12 dogs in the morphine-bupivacaine group, 13 dogs in the morphine group, and 11 dogs in the control group. No significant associations were found among treatments with regard to type of surgery, age, sex, or body weight, suggesting that randomization was effective at achieving homogeneity among treatment groups for these variables.
Temporal data—No significant differences among treatments were observed according to duration of surgery (P = 0.62), duration of anesthesia (P = 0.75), delay between epidural injection and draping (P = 0.12), or delay between epidural injection and recovery (P = 0.17).
Preoperative Evaluation
Behavior and plasma cortisol concentration—No significant differences among treatments were observed on the basis of attitude score (P = 0.96) or cooperation score (P = 0.73). Mean baseline preoperative plasma cortisol concentration was identical among groups (P = 0.20; Figure 1).
Predicted mean plasma cortisol concentrations (nmol/L) at various time points during the first 12 hours after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. a–c = At each time point, different letters indicate significantly (P < 0.05) different values among epidural treatment groups. MB = Morphine-bupivacaine. M = Morphine alone. C = Control (saline [0.9% NaCl] solution). P = Time of preoperative sampling.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean plasma cortisol concentrations (nmol/L) at various time points during the first 12 hours after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. a–c = At each time point, different letters indicate significantly (P < 0.05) different values among epidural treatment groups. MB = Morphine-bupivacaine. M = Morphine alone. C = Control (saline [0.9% NaCl] solution). P = Time of preoperative sampling.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean plasma cortisol concentrations (nmol/L) at various time points during the first 12 hours after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. a–c = At each time point, different letters indicate significantly (P < 0.05) different values among epidural treatment groups. MB = Morphine-bupivacaine. M = Morphine alone. C = Control (saline [0.9% NaCl] solution). P = Time of preoperative sampling.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Intraoperative Evaluation
Rescue analgesia—Survival analysis revealed no differences among protocols for time before intraoperative rescue analgesia was administered (P > 0.11). No association was found between the number of times intraoperative rescue analgesia was administered and treatment (P = 0.24; Table 1).
Mean ± SEM intraoperative end-tidal isoflurane concentrations, number of intraoperative administrations of rescue analgesia, and postoperative pain and manipulation scores in 36 dogs undergoing elective orthopedic surgery that received 1 of 3 epidurally administered analgesic protocols in addition to a parenterally administered opioid premedication.
| Variable | MB | M | C |
|---|---|---|---|
| Minimum iso concentration (%) | 0.70 ± 0.11a | 1.02 ± 0.10b | 1.03 ± 0.11b |
| Time to minimum iso % (min) | 56.52 ± 13.34a | 75.57 ± 11.96a | 61.51 ± 13.44a |
| Maximum iso concentration (%) | 1.17 ± 0.16a | 1.64 ± 0.14b | 1.70 ± 0.16b |
| Time to maximum iso % (min) | 24.41 ± 11.37a | 50.15 ± 10.19a | 37.64 ± 11.45a |
| Intraoperative rescue analgesia (No. of administrations) | 1a | 1a | 4a |
| Postoperative rescue analgesia (No. of administrations) | 0a | 22b | 20b |
| Total pain score (first 12 h) | 1.64 ± 0.56a | 3.08 ± 0.49b | 3.23 ± 0.51b |
| Total pain score (24 h) | 1.37 ± 0.66a | 2.07 ± 0.58a | 1.17 ± 0.59a |
| Manipulation score (first 12 h) | 0.26 ± 0.30a | 1.45 ± 0.26b | 1.69 ± 0.27b |
| Manipulation score (24 h) | 0.73 ± 0.40a | 1.41 ± 0.35a | 0.85 ± 0.36a |
Within each row, different superscripts indicate significantly (P < 0.05) different values.
Iso = Isoflurane. MB = Morphine-bupivacaine. M = Morphine. C = Control (saline [0.9% NaCl] solution).
End-tidal isoflurane concentration—A significant effect of treatment on volatile anesthetic requirement was detected. The minimum end-tidal isoflurane concentration was lower in the morphine-bupivacaine group than in the morphine and control groups (P = 0.015; Table 1), but there was no difference in values between the morphine and control groups. The longer the delay between epidural injection and draping, the lower the minimum end-tidal concentration (P = 0.031); that relationship was similar in all treatment groups (P = 0.87).
A significant (P = 0.008) treatment effect on maximum end-tidal isoflurane concentration was detected (Table 1). The maximum end-tidal concentration of isoflurane was lower in the morphine-bupivacaine group than in the morphine and control groups, and there was no significant difference between the morphine and control groups. An inverse relationship (P = 0.036) was detected in the delay between epidural injection and draping and maximum end-tidal concentration. That relationship was similar in all treatment groups (P = 0.55).
Hemodynamics—A treatment effect on mean arterial pressure was detected (P < 0.001), with values lower in the morphinebupivacaine group than in the morphine and control groups. No differences were observed among groups in minimum mean arterial pressure (P = 0.15); however, there was a marginal effect of treatment group on prevalence of periods of minimum mean arterial pressure. Periods with minimum mean arterial pressure were 8.1 times as likely to occur in dogs in the morphinebupivacaine group as in control dogs (P = 0.017), whereas those periods were equally probable in dogs in the morphine and control groups (P = 0.323) or in dogs in the morphine-bupivacaine and morphine groups (P = 0.099). Three dogs required glycopyrrolate administration (1 in the morphine group and 2 in the control group), and 2 others received dobutamine (both in the morphine-bupivacaine group).
Postoperative Evaluation
Pain score—The effect of treatment on total pain score during the first 12 hours after surgery was significant (P = 0.009; Figure 2; Table 1). Total pain scores were lower in the morphine-bupivacaine group than in the morphine and control groups. There was no difference in pain scores between the morphine and control groups. The effect of treatment on manipulation score during the first 12 postoperative hours was also significant (P < 0.001). The manipulation score changed with time during the postoperative period (P < 0.001; Figure 3). No effect of treatment on preoperative attitude and cooperation scores was detected (P = 0.66 and 0.41, respectively).
Predicted mean total pain scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean total pain scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean total pain scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean manipulation scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean manipulation scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Predicted mean manipulation scores after surgery in dogs undergoing elective orthopedic surgery after receiving 1 of 3 preoperative epidurally administered analgesic treatments in addition to a parenterally administered opioid premedication. Error bars represent SEM. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 229, 7; 10.2460/javma.229.7.1103
Manipulation scores were lower in the morphine-bupivacaine group than in the morphine and control groups, and there was no difference between dogs in the morphine and control groups. Change in the manipulation score with time varied among treatments (P = 0.001). In general, the manipulation score increased with time but did so less quickly in the morphine-bupivacaine group than in the other groups. No difference was detected in total pain or manipulation scores at 24 hours.
Rescue analgesia—Survival analysis revealed no differences among treatments in time before postoperative rescue analgesia was administered (P = 0.92). Contrary to findings in the intraoperative period, an association (P = 0.005) was detected in the postoperative period for the number of postoperative uses of rescue analgesia (Table 1). The number of times postoperative rescue was administered was significantly lower in the morphine-bupivacaine group (n = 0) than in the morphine (22) and control (20) groups, and no difference was detected between the morphine and control groups.
Plasma cortisol concentration—Postoperative plasma cortisol concentration varied significantly with treatment (P = 0.009) and time (P < 0.001; Figure 1). At T0, plasma cortisol concentration was significantly lower in dogs in the morphine-bupivacaine group than in dogs in the morphine and control groups and was significantly lower in dogs in the morphine group than in dogs in the control group. The difference between the morphine-bupivacaine and control groups persisted until T0 + 2 hours, but there was no difference between the morphine-bupivacaine and morphine groups or between the morphine and control groups after T0.
Discussion
Although results of the present study corroborated earlier observations in the literature concerning analgesic efficacy of the epidurally administered morphinebupivacaine combination, the apparent lack of efficacy of morphine alone was surprising. Similar to a curative analgesic protocol in which epidural administration of morphine alone did not induce significantly better analgesia than a saline solution placebo,9 preemptive epidurally administered morphine or saline solution given in addition to a typical clinical premedication analgesic did not induce significantly different degrees of analgesia in dogs of the present study. The morphine-bupivacaine combination was superior at decreasing minimum and maximum end-tidal isoflurane concentrations, the number of doses of postoperative rescue analgesia that were required, pain and manipulation scores during the first 12 hours after surgery, and plasma cortisol concentrations during the initial postoperative period. The synergistic action of opioids and local anesthetic drugs coadministered epidurally or intrathecally has been reported.1–4 It is hypothesized that this synergistic effect explains, at least in part, the better performance of the combination treatment in dogs in the present study. A group of dogs that received epidurally administered bupivacaine alone was not included in the study because it would have increased the number of dogs required for the study without conferring absolute blinded conditions with regard to the investigator performing the postoperative evaluation, and determination of the mechanism of synergistic action of the morphine-bupivacaine combination was not an objective of this study. The literature includes reports on this topic.6,8
In the present study, various aspects of the experimental protocol may have contributed to the apparent poor performance of morphine alone, compared with that of the morphine-bupivacaine combination. The site of action of epidurally administered opioids is not known with certainty. Systemic absorption of drug may account for some effects, as may local diffusion of drug into the spinal cord. Opioids injected into the epidural space can diffuse through the dura mater and penetrate into the dorsal horn of the spinal cord, the anatomic site of nociceptive sensitization. Opioid drugs are thought to work at presynaptic sites in the spinal cord, preventing release of substance P (the neuromediator that initiates the hypersensitizing neuronal wind-up phenomenon), and on postsynaptic receptors to hyperpolarize the cells. Opioids may therefore decrease nociception without having any substantial effect on motor function.
The potency of various opioid drugs administered epidurally has been studied, and it appears that potency is not directly related to systemic potency of the drug, but rather is a function of lipid solubility.15 Morphine has been the most useful opioid for epidural administration because of its high potency and long duration of action (16 to 24 hours). The low lipid solubility of morphine also determines its pharmacokinetics, increasing the time of onset and duration of action when administered by the epidural route. The range in time of onset of action has been reported as 45 to 90 minutes.29,30 Therefore, to achieve the most benefit from an epidural morphine injection, a minimum interval of 45 minutes is required between the time of injection and initiation of the nociceptive stimulation. If surgery begins too soon within that interval, there is risk that analgesia will not have reached optimum effect and efficacy will be partial or null.
The sites of action of local anesthetic drugs administered epidurally are not known with certainty but are thought to be the intradural portion of the spinal nerve roots and periphery of the spinal cord. The time required for onset of motor block after epidural injection of bupivacaine in dogs is approximately 3 to 4 minutes.15 The shorter delay of onset for bupivacaine, compared with morphine, and the importance of inducing effective preemptive analgesia31 may explain the initial perception of superior analgesia associated with the combination. It is possible that the failure of epidurally administered morphine alone to prevent nociceptive sensitization can be explained by the pharmacokinetically mediated long delay, along with or independently of an insufficient analgesic effect induced by opioid receptor stimulation. Sensitization is the result of the exaggerated and uncontrolled influx of nociceptive messages into the dorsal horn of the spinal cord and is associated with an increase in excitability of not only nociceptive neurons, but also of sensitive neurons that typically play no role in pain and nociception (such as the Ab fibers involved in perception of touch, pressure, and vibration).31 Sensitization is also associated with activation of glial cells and a decrease in descending inhibition from the brain.31 Nociceptive sensitization leads to pathologic pain, which exceeds the protective role of pain and is exaggerated and difficult to control. Because preemptive analgesia was induced with oxymorphone or hydromorphone in the anesthetic premedication in the present study, central sensitization was considered unlikely. Whether preemptive analgesic intervention is more effective than conventional regimens in managing acute postoperative pain remains controversial. Preemptive analgesia is a treatment scheme in which the altered sensory CNS processing that amplifies postoperative pain is prevented. The treatment should cover the entire duration of high-intensity noxious stimulation that can lead to establishment of central and peripheral sensitization caused by incision or inflammatory injuries (such as those induced during surgery and the initial postoperative period). Additional analgesic medication will still be needed in the postoperative period, but that pain will be more easily controlled because of the prior administration of preemptive analgesia.
Other practical advantages of preemptive analgesia (via systemic or epidural administration) are that it often decreases the subsequent dose of anesthetic drugs required and that patient safety can be improved concurrently with the effectiveness of pain relief by integrating analgesic therapy into a balanced anesthetic regimen.15 In a recent meta-analysis of 66 studies involving humans,32 the outcomes analyzed were pain intensity, supplemental analgesic consumption, and time to first administration of analgesic. Investigators found a pronounced preemptive effect with epidural analgesia, infiltration of local anesthetics around peripheral nerves, and systemic nonsteroidal anti-inflammatory drug administration. Most impressive in that study were reductions (ranging from 44% to 58%) in supplemental analgesic consumption, at high levels of significance. With systemically acting opioids and N-methyl-D-aspartate receptor antagonists (such as ketamine), the results were equivocal. In a review33 of 80 studies on preemptive analgesia, investigators concluded that there was no clinical value in preemptive analgesia for treatment of postoperative pain; in particular, the use of parenterally administered opioids for preemptive analgesia was not efficacious. This finding may be attributable to development of acute tolerance to opioids, a phenomenon that counteracts the effect of preemptive opioid analgesia whether administered systemically or epidurally. In that scenario, the advantage in preventing surgery-induced sensitization is lost because larger doses of opioids are needed to overcome acute tolerance. This phenomenon of acute tolerance or opioid-induced hyperalgesia has been observed in rodents and humans.34–38 Binding of local anesthetics to sites on voltage-gated Na+ channels prevents opening of the channels by inhibition of the conformational changes that lead to channel activation. This action stops all nervous signals when sufficient numbers of Na+ channels are blocked. By disrupting action potential transmission in all nerve fibers, local anesthetics prevent transfer of noxious information from peripheral nerves or from sites at the level of the spinal cord when administered epidurally. Because these drugs can block all nociceptive input, they are ideal for inducing preemptive analgesia.
Opioids induce analgesia by binding to m, k, and, to a lesser extent, d opioid receptors in the CNS, either in the spinal cord or supraspinally and even at peripheral sites, particularly in inflamed tissue.39 Opioids, like local anesthetics, hinder the electrical activity of neurons by modifying transmembrane ionic conductance. However, in contrast to local anesthetics, opioids have a selective inhibitory effect on nociception without affecting or affecting only minimally other sensory or motor signals. This may explain the differences in analgesic efficacy observed between the morphine and morphine-bupivacaine groups in the present study.
The sensation of pain in animals is estimated indirectly by assessment of pain behavior. Pain behavior (in contrast to pain sensation) has a psychomotor component in addition to a nociceptive component. It is possible that, in animals, the psychomotor component of pain behavior is a dominant factor influencing human assessment of an animal's level of pain and creates bias in the evaluation of analgesic efficacy.37 It could be hypothesized that premedication with hydromorphone or oxymorphone in dogs of the present study obscured the analgesic efficacy of the epidural protocols being tested by 3 possible pathways: decreasing nociceptive sensitization induced by the surgery, inducing behavioral sensitization (leading to psychomotor activity accentuated by the epidurally administered morphine and resulting in falsely positive assessments of the animal's sensation of pain), and causing interference at the level of the opioid receptor with the effect of reducing analgesic efficacy of the epidurally administered morphine (ie, by inducing real tolerance). In the morphine-bupivacaine group, the addition of a local anesthetic to the epidural protocol may have helped inhibit the nociceptive sensitization and opioid tolerance that represent the theoretic basis of multimodal perioperative pain management.35 The increase in intraoperative plasma cortisol concentrations in the morphine and control groups suggested activation of an intraoperative nociceptive barrage sufficient to result in descending hypothalamic-pituitary activity. Local anesthetic blockade of postoperative inflammatory pain inputs significantly attenuated the nociceptive barrage and cortisol release. These findings suggested that maintenance of central sensitization is dependent on input from damaged peripheral tissue, as occurs in the postoperative period. Administered prior to surgery, the epidural morphine-bupivacaine combination yielded better analgesia than morphine alone and should be considered for use in clinical settings.
Ethical concerns precluded studying dogs with no provision of preemptive analgesia, so all dogs in the present study received at least 1 dose of a potent opioid as premedication. Moreover, this experimental design more closely approximated typical clinical situations. Inclusion of a control group in which no analgesic was administered would be a necessary component of a study designed to compare the effects of preoperative analgesic treatment and no treatment. It is known that this approach favors demonstration of a positive effect for the analgesic treatment. For this reason, it is more appropriate to use the term preventive analgesia in that situation and reserve the term preemptive analgesia for the effect of preventive treatment that is limited to sensitization and that begins before surgery but does not extend into the postoperative period. The present study was designed to establish the analgesic efficacy of preemptive epidural analgesia administered under clinical conditions and to compare findings with those from an earlier study,9 in which curative epidurally administered analgesics were tested. Administration of effective preoperative analgesia in all groups attenuated the differences among treatments and may be a reason for the absence of significant differences between the morphine and control groups in some of the variables evaluated in the present study.
Another factor that may have decreased the apparent efficacy of epidurally administered morphine alone was the criteria used to determine the need for postoperative rescue analgesia. The low tolerance for signs of postoperative pain and establishment of a pain score threshold for treatment of only 5 (of 15) were built into the study design and may have attenuated any real differences in efficacy that existed between the treatments in favor of the placebo by requiring more prompt and aggressive intervention.
It is also possible that bias falsely amplified the differences between the morphine-bupivacaine and morphine treatments. Blinding of the evaluator to treatment groups was more difficult to maintain after surgery because of the paraparesis that often results with bupivacaine administration; recognition of that motor block effect could have influenced the evaluator.
The pain scoring system27,28 used was derived from a categoric numeric scale. Use of a variable that specifically evaluated a dog's response to palpation and mobilization of the limb in which surgery had been performed was necessary for differentiating between a high score resulting from nociceptive sensitization and a high score caused by the stress of recovery, dysphoria, or separation anxiety. Several dogs had a total pain score > 5, justifying rescue analgesia, but did not respond to palpation or mobilization of the involved limb (ie, had a manipulation score of 0). The importance of evaluating signs of postoperative pain not only at rest but also during mobilization has been reported in humans.23,24 The effects of different analgesic protocols of unequal efficacy may appear identical during rest but appear different during physical stress. In those 2 human studies,23,24 the effects of epidurally administered morphine-bupivacaine and morphine did not differ in patients at rest, but a difference was evident when the patients mobilized the regions that had undergone surgery. In our study, use of the manipulation score not only revealed a superior analgesic effect for the morphine-bupivacaine combination, but also allowed for detection of differences between the morphine and control protocols. Those differences were not apparent in the total pain scores for the groups, probably because the results of evaluation during mobilization were obscured by less specific results such as vocalization, signs of agitation, or heart rate, which leveled the scores between groups.
The action of local anesthetics in inducing hypotension when administered via the epidural or subarachnoid routes has long been known.11,13,40–42 Strategies that have been implemented to prevent this effect include preliminary loading with IV administered fluids, limiting the maximum volume of drugs administered epidurally and the speed of injection to minimize migration of drug into the cranium and outside the epidural canal (thereby minimizing the effect on the thoracolumbar sympathetic ganglia), diluting the drug to decrease the effect on the type B autonomous nervous fibers, and administering vasopressor drugs. In the present study, bupivacaine administration was associated with lower mean arterial pressures. Nevertheless, the possibility that the lower values for mean arterial pressure were caused by better analgesia instead of depressant hemodynamic effects cannot be ruled out because the conditions of general anesthesia were standardized. Determining the minimum end-tidal isoflurane concentrations permitted objective detection of intraoperative analgesic effects while minimizing the cardiovascular effects of the volatile agent, but it was not possible to ascertain which was the predominant effect. Despite the potential negative hemodynamic effects, the sympatholytic action of local anesthetics may be beneficial in limiting vasoconstriction and decreased tissue perfusion induced by the surgical stress response,30 which has inflammatory and immunologic consequences.43 Although mean arterial pressure was significantly lower in the morphine-bupivacaine group (62.5 ± 2.6 mm Hg) than in the morphine (69.6 ± 2.4 mm Hg) and control (74.8 ± 2.7 mm Hg) groups, the degree of hypotension remained within acceptable limits and was easy to treat with IV administered fluids or dobutamine. The morphine-bupivacaine epidural combination appeared to be safe for use in healthy dogs undergoing elective surgical procedures; however, caution is needed with hemodynamically unstable patients.
In humans44 and other species,45,46 plasma cortisol concentration has been used as a nonspecific indicator of stress and pain. It is known that cortisol secretion has a circadian rhythm independent of stress and pain, with peaks and troughs occurring in a cyclic pattern, although that pattern has been questioned in dogs.46,47 Although it may have been preferable to collect blood samples at the same times each day, the clinical context of the study precluded such standardization. In humans, the intensity of pain varies during the day,48 and circadian variation in the therapeutic index of several drugs has also been reported in animals.49 Nonetheless, most surgeries in dogs of the present study were performed during the morning; thus, the first part of the postoperative period (until T0 + 12) ended at approximately 1:00 to 2:00 AM. Evaluation of dogs for signs of pain was therefore usually conducted during the afternoon and early evening. Results confirmed that plasma cortisol concentrations were higher in dogs with signs of more severe pain. However, this difference was very short-lived, as has been found in other studies.50,w The major differences between treatment groups in plasma cortisol concentrations were detected only during the first 2 to 4 hours of the postoperative period, although differences in apparent severity of pain lasted much longer. These results were in agreement with those of Hendrix et al,9 who confirmed that plasma cortisol concentration is a sensitive marker of response to acute nociceptive input but is not useful in long-term evaluation of pain.
Quandt DJ, Robinson EP, Hendrix PK. Effect of pre-operative epidural lidocaine, morphine, or their combination on isoflurane concentrations in dogs (abstr), in Proceedings. Annu Meet Am Coll Vet Anesthesiol 1998;32.
Numorphan, Dupont Pharma Inc, Mississauga, ON, Canada.
Hydromorphone USP, Sabex Inc, Boucherville, QC, Canada.
Atravet, Ayerst Laboratories Ltd, Montreal, QC, Canada.
Glycopyrrolate USP, Sabex Inc, Boucherville, QC, Canada.
Pentothal, Abbott Laboratories Ltd, Montreal, QC, Canada.
Aerrane, Janssen Inc, Toronto, ON, Canada.
Dobutrex, Novopharm Ltd, Toronto, ON, Canada.
Morphine sulfate HP 25, preservative-free, Sabex Inc, Boucherville, QC, Canada.
Marcaine 0.5%, Sanofi Canada Inc, Markham, ON, Canada.
NaCl 0.9% preservative-free, Astra, Mississauga, ON, Canada.
Pulsator, SIMS Portex Inc, Keene, NH.
Portex nylon epidural catheter, 20G closed end 3 eyes, SIMS Portex Inc, Keene, NH.
Coat-A-Count, Diagnostic Products Corp, Los Angeles, Calif.
Immulite, DPC, Los Angeles, Calif.
Biochem 9100 Multigas Monitor, Roxon Medi-tech Ltd, VilleSt-Laurent, QC, Canada.
Insyte-W 20 GA 1-1, 16 inches, Becton-Dickinson Infusion Therapy Systems Inc, Sandy, Utah.
Pressure monitoring kit with TruWave disposable pressure transducer, Edwards Life Sciences, Baxter Corp, Mississauga, ON, Canada.
Datascope 2000, Datascope Corp, Paramus, NJ.
Datex-Ohmeda model 3800, Datex-Ohmeda, Louisville, Colo.
Nova Stat Profile M, Nova Biomedical, Waltham, Mass.
SAS, version 8.2, SAS Institute Inc, Cary, NC.
Ferreira X, Lambert L, Leblond A, et al. Comparison of analgesia and perioperative problems associated with the pre-operative use of ketoprofen and nimesulide in dogs (abstr). Vet Anaesth Analg 2001;28:207.
References
- 1.
Maves TJ, Gebhart GF. Antinociceptive synergy between intrathecal morphine and lidocaine during visceral and somatic nociception in the rat. Anesthesiology 1992;76:91–99.
- 2.
Penning JP, Yaksh TL. Interaction of intrathecal morphine with bupivacaine and lidocaine in the rat. Anesthesiology 1992;77:1186–1200.
- 3.
Wang C, Chakrabarti MK, Phil M, et al. Specific enhancement by fentanyl of the effects of intrathecal bupivacaine on nociceptive afferent but not on sympathetic efferent pathways in dogs. Anesthesiology 1993;79:766–773.
- 4.
Kaneko M, Saito Y, Kirihara Y, et al. Synergistic antinociceptive interaction after epidural coadministration of morphine and lidocaine in rats. Anesthesiology 1994;80:137–150.
- 5.
Cullen ML, Staren ED, El-Ganzouri A, et al. Continuous epidural infusion for analgesia after major abdominal operations: a randomized, prospective, double-blinded study. Surgery 1985;98:718–727.
- 6.
Akerman B, Arweström E, Post C. Local anesthetics potentiate spinal morphine antinociception. Anesth Analg 1988;67:943–948.
- 7.
Valverde A, Dyson DH, Cockshutt JR, et al. Comparison of the hemodynamic effects of halothane alone and halothane combined with epidurally administered morphine for anesthesia in ventilated dogs. Am J Vet Res 1991;52:505–509.
- 8.
Tejwani GA, Rattan AK, McDonald JS. Role of spinal opioid receptors in the antinociceptive interactions between intrathecal morphine and bupivacaine. Anesth Analg 1992;74:726–734.
- 9.↑
Hendrix PK, Raffe MR, Robinson EP, et al. Epidural administration of bupivacaine, morphine, or their combination for postoperative analgesia in dogs. J Am Vet Med Assoc 1996;209:598–607.
- 10.
Torske KE, Dyson DH, Pettifer G. End tidal halothane concentration and post-operative analgesia requirements in dogs: a comparison between intravenous oxymorphone and epidural bupivacaine alone or mixed with oxymorphone. Can Vet J 1998;39:361–369.
- 11.
Torske KE, Dyson DH, Conlon PD. Cardiovascular effects of epidurally administered oxymorphone and oxymorphone-bupivacaine combination in halothane-anesthetized dogs. Am J Vet Res 1999;60:194–200.
- 12.
Nishiyama T, Hanaoka K. The effects of epidural bupivacaine, morphine, and their combination on thermal nociception with different stimulus intensity in rats. Anesth Analg 2000;91:652–656.
- 13.↑
Troncy E, Junot S, Keroack S, et al. Results of preemptive epidural administration of morphine with or without bupivacaine in dogs and cats undergoing surgery: 265 cases (1997–1999). J Am Vet Med Assoc 2002;221:666–672 (erratum in J Am Vet Med Assoc 2002; 221:1149).
- 14.↑
Lacelles BDX, Waterman AE, Henderson G, et al. Central sensitization as a result of surgical pain: investigation of the pre-emptive value of pethidine for ovariohysterectomy in the rat. Pain 1995;62:201–212.
- 15.↑
Dobromylskyj P, Flecknell PA, Lascelles BD, et al. Management of postoperative and other acute pain. In: Flecknell P, Waterman-Pearson A, eds. Pain management in animals. Toronto, ON, Canada: WB Saunders Co, 2000;81–514.
- 16.
Logas WG, El-Baz N, El-Ganzouri A, et al. Continuous thoracic epidural analgesia for post-operative pain relief following thoracotomy: a randomized prospective study. Anesthesiology 1987;67:787–791.
- 17.
Douglas MJ, McMorland GH, Janzen JA. Influence of bupivacaine as an adjuvant to epidural morphine for analgesia. Anesth Analg 1988;67:1138–1141.
- 18.
Badner NH, Reimer EJ, Komar WE, et al. Low-dose bupivacaine does not improve post-operative epidural fentanyl analgesia in orthopedic patients. Anesth Analg 1991;72:337–341.
- 19.
Badner NH, Komar WE. Bupivacaine 0.1% does not improve post-operative epidural fentanyl analgesia after abdominal or thoracic surgery. Can J Anaesth 1992;39:330–336.
- 20.
Kehlet H, Dahl JB. The value of “multimodal” or “balanced analgesia” in post-operative pain treatment. Anesth Analg 1993;77:1048–1056.
- 21.
Jørgensen H, Fomsgaard JS, Dirks J, et al. Effects of epidural bupivacaine vs combined epidural bupivacaine and morphine on gastrointestinal function and pain after major gynaecological surgery. Br J Anaesth 2001;87:727–732.
- 22.
Torda TA, Hann P, Mills G, et al. Comparison of extradural fentanyl, bupivacaine and two fentanyl-bupivacaine mixtures for pain relief after abdominal surgery. Br J Anaesth 1995;74:35–40.
- 23.
Dahl JB, Rosenberg J, Hansen BL, et al. Differential analgesic effects of low-dose epidural morphine and morphine-bupivacaine at rest and during mobilization after major abdominal surgery. Anesth Analg 1992;74:362–365.
- 24.
Barzoi G, Carluccio S, Bianchi B, et al. Morphine plus bupivacaine vs morphine peridural analgesia in abdominal surgery: the effects on postoperative course in major hepatobiliary surgery. HPB Surg 2000;11:393–399.
- 25.
Valverde A, Dyson DH, McDonell WN. Epidural morphine reduces halothane MAC in the dog. Can J Anaesth 1989;36:629–632.
- 26.
Kiffer F, Joly A, Wodey E, et al. The effect of pre-operative epidural morphine on post-operative analgesia in children. Anesth Analg 2001;93:598–600.
- 27.
Pibarot P, Dupuis J, Grisneaux E, et al. Comparison of ketoprofen, oxymorphone hydrochloride, and butorphanol in the treatment of postoperative pain in dogs. J Am Vet Med Assoc 1997;211:438–444.
- 28.
Grisneaux E, Pibarot P, Dupuis J, et al. Comparison of ketoprofen and carprofen administered prior to orthopedic surgery for control of postoperative pain in dogs. J Am Vet Med Assoc 1999;215:1105–1110.
- 29.
Valverde A, Dyson DH, Mc Donell WN, et al. Use of epidural morphine in the dog for pain relief. Vet Comp Orthop Traumatol 1989;2:55–58.
- 30.↑
Troncy E. Epidural analgesia in the dog and the cat [in French]. Le Médecin Vétérinaire du Québec 1998;28:24–32.
- 31.↑
Muir WW III, Woolf CJ. Mechanisms of pain and their therapeutic implications. J Am Vet Med Assoc 2001;219:1346–1356.
- 32.↑
Ong CK-S, Lirk P, Seymour RA, et al. The efficacy of preemptive analgesia for acute postoperative pain management: a meta-analysis. Anesth Analg 2005;100:757–773.
- 33.↑
Moiniche S, Kehlet H, Dahl JB. A qualitative and quantitative systematic review of preemptive analgesia for postoperative pain relief. Anesthesiology 2002;96:725–741.
- 34.
Kissin I, Lee SS, Arthur GR, et al. Time course characteristics of acute tolerance development to continuously infused alfentanil in rats. Anesth Analg 1996;83:600–605.
- 35.↑
Ossipov MH, Lai J, King T, et al. Antinociceptive and nociceptive actions of opioids. J Neurobiol 2004;61:126–148.
- 36.
Simonnet G. Opioids: from analgesia to anti-hyperalgesia? Pain 2005;118:8–9.
- 37.↑
Kissin I. Tolerance to opioid analgesia: why do we differ from rats? Anesth Analg 2005;101:1727–1729.
- 38.
Angst MS, Clark JD. Opioid-induced hyperalgesia: a qualitative systematic review. Anesthesiology 2006;104:570–587.
- 39.↑
Branson KR, Gross ME. Opioid agonists and antagonists. In: Adams HR, ed. Veterinary pharmacology and therapeutics. 8th ed. Ames, Iowa: Iowa State University Press, 2001;268–298.
- 40.
Persson F. Epidural analgesia in dogs with special reference to the intraarterial blood pressure. Acta Vet Scand 1970;11:186–196.
- 41.
Nolte I, Watney GCG, Hall LW. Cardiovascular effects of epidural blocks in dogs. J Small Anim Pract 1983;24:17–21.
- 42.
Franquelo C, Toledo A, Manubens J, et al. Bupivacaine disposition and pharmacologic effects after intravenous and epidural administrations in dogs. Am J Vet Res 1995;56:1087–1091.
- 43.↑
Kona-Boun JJ, Silim A, Troncy E. Immunologic aspects of veterinary anesthesia and analgesia. J Am Vet Med Assoc 2005;226:355–363.
- 44.↑
Domzal T, Szczudlik A, Kwasucki J, et al. Plasma cortisol concentrations in patients with different circadian pain rhythm. Pain 1983;17:67–70.
- 45.
Smith JD, Allen SW, Quandt JE, et al. Indicators of postoperative pain in cats and correlation with clinical criteria. Am J Vet Res 1996;57:1674–1678.
- 46.
Smith JD, Allen SW, Quandt JE. Changes in cortisol concentration in response to stress and postoperative pain in client-owned cats and correlation with objective clinical variables. Am J Vet Res 1999;60:432–436.
- 47.
Kemppainen RJ. Principles of glucocorticoid therapy in nonendocrine diseases. In: Kirk RW, ed. Current veterinary therapy: small animal practice IX. Philadelphia: WB Saunders Co, 1986;954–962.
- 48.↑
Glynn CJ, Lloyd JW. The diurnal variation in perception of pain. Proc R Soc Med 1976;69:369–372.
- 49.↑
Moore Ede MC. Circadian rhythms of drug effectiveness and toxicity. Clin Pharmacol Ther 1973;140:925–935.
- 50.↑
Grøndahl-Nielsen C, Simonsen HB, Damkjer Lund J, et al. Behavioral, endocrine and cardiac responses in young calves undergoing dehorning without and with use of sedation and analgesia. Vet J 1999;158:14–20.
Appendix 1
Scoring system used to calculate preoperative attitude and cooperation scores in 36 dogs undergoing elective orthopedic surgery that received 1 of 3 epidurally administered analgesic protocols in addition to a parenterally administered opioid pre-medication.
| Behavioral variables | Score |
|---|---|
| General attitude | |
| Apathetic or indifferent | 1 |
| Friendly | 2 |
| Nervous, submissive | 3 |
| Very nervous, tries to move away | 4 |
| Aggressive | 5 |
| Cooperation during venipuncture | |
| No objection | 0 |
| Recognizes injection, no complaint | 1 |
| Objects, but does not try to bite | 2 |
| Tries to bite, struggles violently | 3 |
Appendix 2
Scoring system used to evaluate the same dogs for postopera tive pain.
| Variable | Score |
|---|---|
| Vocalization | |
| No crying | 0 |
| Crying, responds to calm voice | 1 |
| Crying, does not respond to calm voice | 2 |
| Agitation | |
| Asleep or calm | 0 |
| Mild agitation | 1 |
| Moderate agitation | 3 |
| Severe agitation | 4 |
| Respiration | |
| Normal effort | 0 |
| Mild abdominal assistance | 1 |
| Marked abdominal assistance | 2 |
| Heart rate | |
| <10% increase relative to the preoperative value | 0 |
| 10%–30% increase relative to the preoperative value | 1 |
| 31%–50% increase relative to the preoperative value | 2 |
| >50% increase relative to the preoperative value | 3 |
| Manipulation of the surgical wound* | |
| No response | 0 |
| Minimal response, tries to move away | 1 |
| Turns head toward site, slight vocalization | 2 |
| Turns head toward site, loud vocalization | 3 |
| Turns head with intention to bite, howls | 4 |
The score for this item was called the manipulation score and was used to differentiate signs of dysphoria from signs of true pain. Scores were determined on the basis of the dog's response to stimuli of increasing intensity, including gentle pressure on the surgical wound and gentle flexion and extension of the affected joint.
Appendix 3
Statistical tests and comparisons and variables evaluated in the same dogs that were assigned pre-and postoperative pain and manipulation scores by use of Appendices 1 and 2.
| Variables | Type of variable | Statistical tests |
|---|---|---|
| Age | Quantitative continuous | Linear model |
| Body mass | ||
| Duration of surgery | ||
| Duration of anesthesia | ||
| Delay epidural-draping | ||
| Delay epidural-recovery | ||
| Minimum end-tidal isoflurane concentration | ||
| Time required to reach minimum end-tidal isoflurane concentration | ||
| Maximum end-tidal isoflurane concentration | ||
| Time required to reach maximum end-tidal isoflurane concentration | ||
| Intraoperative MAP | ||
| Minimum MAP | ||
| Preoperative cortisol | ||
| Total pain score at 24 hours | Qualitative ordinal | |
| Manipulation score at 24 hours | ||
| Postoperative cortisol | Quantitative continuous | Linear model with repeated measures |
| Total pain score during the first 12 postoperative hours | Qualitative ordinal | |
| Manipulation score during the first 12 postoperative hours | ||
| Score of attitude | Qualitative ordinal | Exact χ2 |
| Score of cooperation | ||
| Sex | Qualitative binary | |
| Administration of intraoperative analgesia | ||
| Administration of postoperative analgesia | ||
| Type of surgery | Qualitative multiple | |
| Delay for the use of intraoperative analgesia | Quantitative continuous | Survival analysis (Cox model) |
| Delay for the use of postoperative analgesia | ||
| Total number of analgesic doses | Quantitative discontinuous | Negative binomial regression |
| Prevalence of minimum MAP |
MAP = Mean arterial pressure.
