Introduction

Postoperative pain has been associated with several complications and delayed recovery to normal function.¹ Drugs used for treating acute pain include opioids, nonsteroid anti-inflammatory drugs, local anesthetics, α2-adrenergic agonists, and N-methyl-D-aspartate receptor antagonists.^1–3

Mu opioid receptor agonists, such as morphine, are considered the most efficacious group of analgesic drugs. They give dose-dependent analgesia with minimal side effects, such as bradycardia, respiratory depression, nausea, dysphoria, and opioid tolerance, at clinically useful doses.^1,4,5 Constant rate infusions (CRIs) of morphine at a dose of 0.12 mg/kg/hr showed an equianalgesic effect to 1 mg/kg intramuscular boluses every 4 hr in dogs undergoing laparotomy when assessed with a behavioral composite pain scale.⁴ However, morphine CRI of 0.15 mg/kg/hr showed less mechanical antinociception, assessed with von Frey filaments, than 0.5 mg/kg IV boluses every 2 hr.⁶

In human medicine, the local anesthetic lidocaine has been proven to block ectopic discharges from injured neurons both peripherally and spinally; to reduce stress response, platelets aggregation, and leukocytes activation in trauma patients; and is known to have antioxidant properties.⁵ Constant rate infusions of lidocaine ranging from 15 to 400 mcg/kg/min produces dose dependent minimal alveolar concentration (MAC) sparing effects while 25 mcg/kg/min lidocaine CRI produced analgesic effects comparable to those of a morphine CRI after phacoemulsification in dogs.^7–10 However, 12 hr lidocaine CRI with doses ranging from 10 to 100 mcg/kg/min failed to reduce mechanical nociceptive threshold in awake dogs.¹¹

Ketamine is an injectable, dissociative anesthetic with analgesic properties attributed to prevention of N-methyl-D-aspartate receptor mediated central sensitization.¹² However, the complex mechanism of action may involve opioid synergism and prevention of opioid tolerance, anti-inflammatory, antitumoral, antishock, and neuroprotective properties.^2,13 Perioperative administration of 2.5 mg/kg ketamine intramuscular boluse decreased rescue analgesia and visual analogue pain score in bitches undergoing ovariohysterectomy.¹⁴ Even though 0.5 mg/kg ketamine IV bolus followed by 10 mcg/kg/min CRI failed to reduce nociceptive withdrawal response (NWR) in conscious dogs, perioperative administration of ketamine CRI significantly reduced pain scores in dogs undergoing forelimb amputation and reestablished normal feeding behavior in bitches undergoing mastectomy.^3,15,16

Based on individual and combined actions of the components, morphine-lidocaine-ketamine constant rate IV infusion (MLK) is currently advocated as a means for providing postoperative analgesia. Although morphine, lidocaine, and ketamine alone and combined can reduce the MAC of inhaled anesthetics, MAC reduction observations cannot predict or evaluate effectiveness in providing post-operative analgesia.^9,17 Inhaled anesthetic MAC reduction studies are performed on unconscious animals or humans.¹⁸ Analgesia is defined as the absence of pain, and pain is defined as an “unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage.”¹⁹ As such, the perception of pain is a conscious experience. Thus, in order to objectively evaluate analgesic effectiveness of multimodal analgesia combinations, subjects must be capable of experiencing pain. Aside from a recent study comparing single dose of IM morphine, MLK and epidural postoperative analgesia in dogs undergoing tibial plateau levelling osteotomy, there are no published scientific veterinary reports supporting the use of MLK for postoperative analgesia.¹⁷

The objective of the present study was to evaluate and compare antinociceptive effect of morphine constant rate IV infusion (M), morphine-lidocaine constant rate IV infusion (ML), or MLK in incision-induced pain in dogs. We hypothesized that MLK administered as a CRI for 18 hr postsurgery would show a more profound analgesic effect than the other two combinations.

Materials and Methods

The study was performed following the guidelines of the Laboratory Animal Resources and Care at Mississippi State University and the protocol was approved by the Institutional Animal Care and Use Committee (protocol number 13-024).

Animal Care and Instrumentation

Eighteen adult mixed-breed dogs, weighing between 18 and 36.3 kg, were used in this study. Dogs were part of a project comparing barbed suture device and conventional monofilament suture for intradermal skin closure. All dogs were considered healthy on the basis of physical exam, auscultation, complete blood count, chemistry, and electrocardiogram. Dogs were randomly assigned to three different treatments using a random list created with commercial software^a. The treatments consisted of the following IV infusions: 5 mL/kg/hr of electrolyte solution^b containing 0.024 mg/mL of morphine sulfate^c (0.12 mg/kg/hr) (M); 5 mL/kg/hr of electrolyte solution containing 0.024 mg/mL of morphine sulfate and 0.6 mg/mL of lidocaine hydrochloride^d (3 mg/kg/hr) (ML); 5 mL/kg/hr of electrolyte solution containing 0.024 mg/mL of morphine sulfate, 0.6 mg/mL of lidocaine hydrochloride, and 0.12 mg/mL of ketamine hydrochloride^e (0.6 mg/kg/hr) (MLK).

Food, but not water, was withheld for approximately 12 hr before the beginning of the study. The morning prior to surgery, dogs were clipped in the thoracic region and a 6 cm line was drawn with an indelible marker at the predicted surgical site for the suture study; then a rectangular pattern was drawn around the line with the help of a ruler at 2, 5, and 10 mm spacing from the drawn line for baseline assessments. The right cephalic vein was catheterized percutaneously with an 18 gauge 2 inch IV catheter^f and all dogs were deeply sedated with a combination of dexmedetomidine^g (0.01 mg/kg) and morphine sulfate (0.2 mg/kg) IV. Lidocaine 2% (0.2 mL) was applied locally on the arytenoids prior to endotracheal intubation with a cuffed polyvinyl chloride endotracheal tube with a Murphy's eye. Oxygen was supplied at 2 L/min through a Universal F breathing circuit. In the event the procedure lasted longer than 30–45 min or sedation was judged insufficient, additional doses of 0.0025 mg/kg of dexmedetomidine IV were administered. In case of purposeful movement, the dog was administered 1 mg/kg of propofol^h bolus IV. Monitoring included oscillometric arterial blood pressure, electrocardiogram, pulse oximetry, and capnographyⁱ. As part of the study comparing different suture materials, all animals received 6 skin incisions (6 cm inch length each), three on each side of thoraco-lumbar area, that were randomly sutured using either novel or conventional suture. Three rectangular patterns were drawn around the most caudal incision on the right side of the dog as previously described, for assessment of mechanical nociception with von Frey filaments (Figure 1). At the end of the procedure, dexmedetomidine was reversed with atipamezole^j (0.1 mg/kg intramuscular). Dogs were extubated following return of the swallowing reflex, at which time loading doses of lidocaine (2 mg/kg) and ketamine (0.5 mg/kg) or lidocaine (2 mg/kg) and saline or equal volume of saline were administered IV by an operator (Ludovica Chiavaccini) blinded to the treatment. Loading dose of morphine was not considered necessary because the infusion was started within an hr from morphine premedication. The corresponding infusion was started after extubation and continued for 18 hr using a peristaltic infusion pump^k. All dogs were restrained in kennels during the infusions.

View Figure

• Download as Powerpoint
• Download Figure

Measurements

All measurements were performed by a single observer (Ludovica Chiavaccini) who was blinded to the treatment prior to sedation, at extubation, and 1, 4, 8, 12, 24, and 48 hr after extubation. Sedation was scored using a simple descriptive scale 0 to 2 (Table 1). Nociception was estimated using in sequence the Short Form Glasgow Composite Measure Pain Scale (GCMPS-SF) and variably sized von Frey filaments^l. The von Frey filaments threshold measurement was performed as follows: filaments were used in ascending order, starting with the smallest (1.65) and progressing to stiffer filaments, until a response was noted. Each filament was applied at three different points of each side of the rectangle 2 mm from the incision. The filament was applied until it bowed. If no response was obtained, the diameter of the filament was increased until the dog showed a purposeful response (turning toward the stimulus or withdrawing from the stimulus). If no response was elicited, the highest stiffness of filament (6.65, corresponding to a 300g target force) was recorded. The procedure was repeated for rectangles 5 and 10 mm from the incision. The von Frey filaments score (VFFS) was calculated for each time interval by adding together the von Frey threshold measurement (expressed in grams) obtained at 2, 5, and 10 mm from the incision.

TABLE 1 Simple Descriptive Sedation Score

View Fullscreen

Results

Of the 18 dogs enrolled, 9 were females and 9 were males. Their median weight was 31 kg (24–32 kg), 29.5 kg (19–34 kg), and 24 kg (19–33 kg) for the MLK, ML, and M groups, respectively. There was no significant difference in gender distribution (p = .83) or weight (p = .58) across the three groups. At extubation, 5/18 (28%) dogs recorded a sedation score 0, 7/18 (39%) recorded sedation score 1 and 6/18 (33%) sedation score 2; however, there was no difference in sedation score between groups over time (p = .31–.95). The GCMPS-SF and the VFFS over time in the MLK, ML, and M groups are presented in Figures 2 and 3. There was no difference in GCMPS-SF between groups (p = .57); however, the score was significantly higher than baseline between extubation and 24 hr postoperative (p < .0001). None of the dogs reached the threshold GCMPS-SF set for administration of rescue analgesia. Even though sedation was positively associated with GCMPS-SF (p = .0001), this difference lost significance when considered over time. Dogs in the ML treatment group had significantly lower VFFS (p = .02) than dogs in the M group at all times, while there was no difference between M and MLK treatments (p = .78). However, the VFFS was significantly lower than baseline between extubation and 48 hr postoperative (p < .0001). While holding all other predictors in the model fixed, sedation score was positively associated with VFFS (p < .0001). Eleven (61%) dogs showed sialorrhea within the first 12 hr of infusion with the majority of the episodes happening between 1 and 4 hr (73%); however, there was no difference in incidence of sialorrhea between groups (p = 1.0) (Table 2). There was no association between salivation and grade of sedation at any given time point (p = .18). None of the dogs showed other signs of nausea such as lip licking and retching or emesis.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

TABLE 2 Number (Proportion) of Dogs Showing Hypersalivation During 18-hr Infusion of Morphine Alone, Morphine-Lidocaine, and Morphine-Lidocaine-Ketamine

View Fullscreen

Discussion

Results from this study suggest that in an incision-induced pain model, there is no difference in mechanical nociceptive response to von Frey filaments between MLK and M. ML infusion, on the other hand, seems to lower the response threshold with respect to morphine alone. No difference was observed in GCMPS-SF between the three treatments.

Pain is a multidimensional condition and multimodal analgesia has been proved more effective than using a single drug.^1,5 However, there is little quantitative information on postoperative constant infusions using multimodal analgesics in dogs. Wagner et al. compared postoperative IV infusions of fentanyl alone (1–5 mcg/kg/hr) to a combination of fentanyl and ketamine (10 mcg/kg/min during surgery and 2 mcg/kg/min for 18 hr after) in dogs undergoing forelimb amputation.¹⁶ While ketamine did not reduce postoperative analgesic requirements, dogs receiving perioperative infusion of ketamine showed better demeanor and increased activity by the third day after surgery. Similar findings were obtained in a later study, in which postoperative intravenous administration of 0.7 mg/kg ketamine followed by 6-hr ketamine CRI at 10 mcg/kg/min improved the food intake of female dogs within 8 hr after mastectomy, without, however, reducing opioid requirements.³ The analgesic effect of ketamine is controversial. In an experimental rat model and in humans, subanalgesic doses of ketamine seemed to potentiate the opioid effect and reduce opioid tolerance within 24 hr after surgery.^2,13 A combination of morphine (0.1 mg/kg) and ketamine (0.1 mg/kg followed by 4 mcg/kg/min CRI) reduced the stimulus-response curve to NWR in 12 conscious human volunteers, but it did not reduce the painful sensation or the wind-up phenomenon.²⁰ On the other hand, ketamine CRI alone failed to reduce NWR or behavioral response in awake dogs after single transcutaneous electrical stimulation.¹⁵ Potential reasons for these discrepancies could be either that ketamine mainly acts on the development of central sensitization and hyperalgesia or that the effective dosing regimen of ketamine CRI has not been defined yet.^12,15 Nonetheless, MLK infusion has made inroads in veterinary practice.^17,21 Interestingly, we did not find any difference in the mechanical nociceptive threshold measured with either the von Frey filaments or the GCMPS-SF between M and MLK treatments.

There were only two studies evaluating the postoperative analgesic effects of MLK infusion at the time of writing. The first one is a case report documenting the postoperative management of a 8 yr old springer spaniel undergoing forelimb amputation.²² In that study, bradycardia, altered mentation, dysphoria, and panting were reported during infusion. The sole side effect we observed was sialorrhea, which occurred in 61% of dogs. Morphine is known to cause nausea and vomiting in dogs after intramuscular injection and CRI.^4,23 Nausea and emesis have also been reported in dogs receiving 12-hr lidocaine infusion at 50 mcg/kg/min or more.¹¹ We did not observe any difference in incidence of sialorrhea between treatments or any association between sialorrhea and sedation over time, so this side effect was most likely due to morphine infusion. The second one is a clinical trial comparing MLK, epidural morphine and ropivacaine, and both treatments together to sole morphine intramuscular premedication in 48 dogs undergoing tibial plateau levelling osteotomy.¹⁷ The authors found no difference in postoperative pain scores, sedation scores, rescue analgesia requirements, or time to first rescue analgesia between groups.¹⁷

Pain/nociception assessment in animals can be difficult without a validated and reliable objective scale.^17,24 In the present study, we used two different outcome measures of nociception, the VFFS and the GCMPS-SF. Von Frey filaments elicit mechanical nociceptive stimulus based on a range of applied pressures. They have been used in incisional models in several animal species without causing intolerance, learned aversion, or local hyperalgesia over time.^23,25 Usually an area remote from the incision is used as control.^23,25 Our study tested the surgical area prior to sedation and incision instead. The VFFS was significantly lower than baseline in all groups for 48 hr after extubation. There was no difference in VFFS between MLK and M groups, whereas the ML group showed significantly lower VFFS than the M group at all times. We couldn't find any pharmacological explanation for this discrepancy. This group of dogs recorded an overall lower VFFS at baseline too, even if the difference was not statistically significant, which may indicate that the ML group was more sensitive to mechanical pressure applied on the skin and the difference may have been exacerbated by the large individual variability, the small sample size, and surgery.

As von Frey threshold measurement requiring rescue analgesic intervention has not been defined in dogs yet, we chose to use the short form of the Glasgow pain score as reference for rescue analgesia.²³ The GCMPS-SF is a specific score based on six behavioral categories with scaled answers and a total score ranging between 0 (less painful) and 24 (maximally painful).²⁶ It was validated in English for dogs undergoing ovariohysterectomy, but it has been successfully applied in facilities where English was not the first language and its use in canine studies involving procedures other than ovariohysterectomy has been documented.^17,27 The GCMPS-SF was significantly higher than baseline in all groups for 24 hr after extubation, with no statistically significant difference between treatments. One of the main limitations of behavioral pain scales is that they do not leave room for adjustment for non-nociceptive stressors and they do not account for individual temperament and psychological status of the animal.^17,25 Although the experimental area was kept quiet and separate, the dogs were not well socialized and some of them were fearful during examination. The Glasgow Composite Measure Pain Scale is known to be less affected by sedation than other pain scoring systems; however, in order to control for sedation as a possible confounding variable, we entered it as covariate into the final multivariable regression models.^17,24 In addition, due to ethical considerations, all dogs received morphine infusion. Morphine is an effective analgesic and is likely to have masked the effects of the other drugs in this model of minimal to moderate pain. While the GCMPS-SF was not sensitive enough to discriminate analgesic differences between treatments in the current model, different results may have been obtained with a model involving more intense or prolonged pain. However, none of the dogs reached the threshold score for rescue analgesia.

The main weakness of the present study was the small sample size. The R² of the model was obtained by squaring the correlation between the observed response and the predicted response and it was confirmed through predictor correlations. Power analysis was developed using a free, stand-alone programⁿ. Assuming a two-tailed multivariable mixed effect linear regression model, with H0 R²=0 and α=0.05, H1 R²=0.36 (GCMPS-SF) or 0.23 (VFFS) and 9 (GCMPS-SF) or 11 (VFFS) predictors, 45 dogs and 80 dogs would have been required to yield a power of 0.8 in testing hypotheses concerning the three categorical research variables: treatment, sedation, and time. We did not have control over the sample size since the dogs used in this study were part of another study comparing two different types of skin suture materials. The absence of a negative control group limited our possibility to make any inference on the absolute analgesic effect of any of the combinations. However, the aim of the study was to compare the relative efficacy of ML and MLK to M. The third weakness is that the combination of morphine, lidocaine, and ketamine in one bag and the use of a peristaltic pump can reduce the precision of drug amount delivered. Nonetheless, we aimed to recreate the protocol commonly recommended for use in the clinical setting.²¹ In addition, two types of suture were used in this study; the sutures were made of the same material and only structurally different (barbed versus not barbed). The barbed suture is expected to be less irritating due to more equal distribution of tension along the incision line and less suture material being left in the patient due to the absence of a knot at both ends. One could argue that this might have affected the results; however, all dogs received the same number of barbed versus nonbarbed suture in random order. The investigator was blinded to which suture was used. Last but not least, previous studies showed that the beneficial effects of ketamine may be delayed with respect to its administration. Increasing the duration of data collection in this study may have better shown these effects.

Conclusion

In conclusion, we found no significant analgesic difference in conscious dogs between infusion of MLK or M alone using an incision-induced pain model and VFFS or GCMPS-SF for assessment. Even if perioperative ketamine could have some beneficial effects, these effects are probably relevant only in surgeries associated with more severe and prolonged pain. Infusion of a combination of morphine and lidocaine was associated with a lower VFFS at all times, but further studies are needed to better understand if this combination is consistently less efficacious than morphine alone or these findings were biased by the small sample size.