The glutamate-activated N-methyl-D-aspartate (NMDA) receptor plays a crucial role in the development of central sensitization. Some of the characteristics associated with central sensitization are manifested clinically by a painful sensation from a non-painful stimulus (allodynia) and augmented pain sensation from previously painful stimuli (hyperalgesia) [1–5].

Association of the NMDA receptor with postoperative pain ushered a new chapter in the exploration of NMDA-antagonists for postoperative analgesia protocol. Ketamine is well established NMDA- antagonist. It is commonly used as an anaesthetic agent in children and animals. It is useful for its anti-nociceptive properties in acute pain, and can be used for chronic and neuropathic pain as well [4]. Ketamine has prevented hyperalgesia, improved analgesia [6–8] and decreased postoperative opioid consumption [8–10] in surgical patients.

The analgesic dose range of ketamine for use in dogs has not been properly established. As a consequence, analgesic benefits of ketamine are not yet fully recognized in dogs. The effective dose range of ketamine in human surgical patients has been established at a loading dose of 0.5 mg/kg, followed by continuous rate infusion (CRI) of 2 to10 μg/kg/min [7, 10–12]. In the absence of appropriate analgesic dose range, the doses used in studies of dogs [13–15] have been taken from human studies [16, 17]. This has resulted in reporting inconclusive results. Moreover, most of the studies were clinical. Consequently, the use of opioids pre and postoperatively in these studies might have masked the analgesic effects of ketamine. Apart from this, various scoring scales used in these studies have not been validated for reliability, specificity or linearity. Most importantly these scales were based on subjective assessment of behaviour [18, 19], which can be biased by inter-observer variability. Furthermore, none of these studies correlated effects with serum concentrations of ketamine.

So far, only one study reported plasma concentration versus antinociceptive actions in dogs [20]. In this study, ketamine was administered at 0.5 mg/kg, followed by 10 μg/kg/min for 59 min in dogs, antinociceptive effect was demonstrated only for the first four minutes, at which time the plasma concentration was >100 ng/mL. Plasma concentration of ketamine declined to <100 ng/mL after 5 min [20]. Further analysis revealed that the plasma concentrations of ketamine at pseudo-steady state were five times lower than the expected concentrations to that in humans receiving the same CRI. They concluded that this dose was insufficient to maintain the analgesic concentration of the ketamine. This also suggests that ketamine pharmacokinetics in dogs may differ from that in humans, and that may warrant the use of higher infusion rates in dogs.

Lidocaine is commonly used as a local anaesthetic and anti-arrhythmic agent in veterinary and human medicine. With the mechanism based approach combination of lidocaine acting on sodium channels, which play a part in generation and transmission of nociceptive impulses [1] and ketamine acting on NMDA receptors would be better strategy to manage postoperative pain, compared to conventional method of administering single drug for post operative pain management. The major benefit of combining lidocaine and ketamine is to prevent the development of central sensitization during surgical intervention [21], which may reduce pain and discomfort in the postoperative period. The combination of ketamine and lidocaine has been reported to decrease the minimum alveolar concentration (MAC) of inhalant anaesthetics in dogs [22–25]. It has been suggested that the MAC reducing effect may be due to analgesic or sedative effect or both [22, 24, 25]. The analgesic effects of the combination of lidocaine and ketamine in conscious dogs have not been reported.

This study therefore, aimed to correlate the serum concentrations of ketamine to the antinociceptive effects using two infusion rates of ketamine. Furthermore, this study also evaluated the clinical and antinociceptive effects of incorporation of lidocaine infusion to ketamine in a controlled, repeatable situation.

The experiments were well tolerated by all the dogs. Changes in behavior due to recognized psychomimetic effects of ketamine were mild and harmless. The behavioral changes disappeared within 20 min after the end of infusion, and the dogs resumed their activity immediately after the end of experiment. The algometer device was well tolerated by all the animals. Transient dimples were observed at the carpal pad, metacarpal footpad, and abdomen, which disappeared within 1 min, before the replicate measurement. There was no sign of redness, swelling, bleeding, exudates, and lameness during, immediately after, and 24 h after the experiment.

Serum concentrations of ketamine

The serum concentrations (Mean ± SD) of K30, KL30 and K50 across time points are illustrated in Fig. 1. Maximum concentration was observed at 1 min for K30 (435.34 ± 26.18 ng/mL), KL30 (582.34 ± 227.46 ng/mL), and K50 (733.77 ± 133.6 ng/mL), which remained stable for 20 min, and declined gradually reaching around 250.87 ± 39.87, 221.73 ± 91.03 and 343.67 ± 63.21 ng/mL at 120^th min in K30, KL30 and K50, respectively. The serum concentration decreased rapidly to less than 100 ng/mL within 20 min following the end of infusions in all three treatments. Between 1 and 40 min, serum concentration of K50 was higher than K30, while concentration of KL30 lied in between K50 and K30. By 60^th min, the three groups were not statistically different.

Nociceptive mechanical thresholds

There was no significant difference among the baseline mechanical thresholds at carpal pad, metacarpal foot pad, tibia, femur and abdomen in all treatment groups (Fig. 2a-e). All the three treatments increased thresholds significantly (P < 0.05) throughout the infusion up to 120 min. Mechanical thresholds returned to baseline within 20 min of the end of CRI at 140^th min. Overall, nociceptive mechanical thresholds were higher in KL30 than in K30 and also higher in K50 than in KL30, and some significant differences between groups were recorded. There was weak positive correlation between serum concentration of K30 with mechanical thresholds at carpal pad (r-0.24 P = 0.141), metacarpal footpad (r-0.32, P = 0.053), tibia (r-0.33, P = 0.047), femur (r-0.30, P = 0.075) and abdomen (r-0.35, P = 0.053). Likewise, weak positive correlation of K50 with mechanical thresholds at carpal pad (r-0.29 P = 0.080), metacarpal footpad (r-0.38, P = 0.020), tibia (r-0.18, P = 0.330), femur (r-0.22, P = 0.185) and abdomen (r-0.12, P = 0.527) were observed.

The concentrations of ketamine in the present study were 435.34 ± 26.18 ng/mL for K30, 582.34 ± 227.46 ng/mL for KL30 and 733.77 ± 133.6 ng/mL for K50 at 1 min and 250.87 ± 39.87, 221.73 ± 91.03 and 343.67 ± 63.21 ng/mL at 120 min for K30, KL30 and K50, respectively. The serum concentrations of ketamine produced by all the 3 CRI regimens in this study could be maintained above 100 ng/mL during the 120- min infusions. Furthermore, nociceptive thresholds at most body points were raised above baseline values during the 120 min of infusions. The nociceptive thresholds returned towards baseline values within 20 min after cessation of infusions, at which point, serum levels were found to be lower than 100 ng/mL. Results from this study suggested that the minimum concentration of ketamine to produce analgesia in dogs is between 100 as observed at 140 min and 200 ng/mL during infusion. This concurred with findings in humans, where serum concentration of 100–200 ng/mL of ketamine was reported to be analgesic with minimal side effects [11, 31].

Effective analgesic concentration of ketamine has not been reported in dogs. The only study reporting the analgesic concentration of ketamine versus its antinociceptive action in dogs has been conducted by Bergadano et al. [20]. They found that the analgesic effect only lasted for the first four minutes at plasma concentration of 220–370 ng/ mL. No further analgesic effect was observed thereafter for up to 59 min, when plasma concentration was found to be below 100 ng/mL. They selected a ketamine dose of 0.5 mg/kg and 10 μg/kg/min from the human studies [16, 17, 31, 32]. However, these dosages had difficulties to sustain a desired plasma ketamine concentration (i.e. above 100–200 ng/mL) to produce antinociceptive effect conscious dogs. The authors suggested that this variation was due the difference in pharmacokinetics of ketamine between human and dogs, in addition to inter-species an intra species difference has also been reported in dogs [33]. Various factors could contribute to the individual variations within the species; these include age, body weight, lean-to-fat body ratio and genetics [34–36]. Although there is no data reporting the influence of genetic factors on the disposition of ketamine in dogs, its potential role cannot be excluded. In this study, we selected the ketamine dose based on the prior results reported by Bergadano et al. [20] . We used similar loading dose of 0.5 mg/kg, but increased the CRI to 30 and 50 μg/kg/min, doses 3 and 5 times higher than that used by Bergadano. Both the doses maintained serum concentration above the level required to raise the mechanical nociceptive thresholds up to 120 min. Our study showed that the mechanical nociceptive threshold in dogs can be raised through ketamine infusion. In order to maintain this plasma concentration and have an antinociceptive effect over 120 min, higher infusions rates than previously employed by Bergadano et al. in dogs (20) are deemed necessary. A time of 120 min was selected in our study to simulate a multimodal analgesic approach that could provide pre-intra and post-operative antinociception for a potential surgery of 45 min. This period covers the time before, during and after surgery up to recovery. Thus, results of this study confirmed the need for higher infusion rates of ketamine in dogs than that previously employed by Bergadano in dogs [20].

In the present study, the concentration of ketamine was higher during first 10–20 min of adminsitration, and declined gradually thereafter. The initial hike and maintenance of the higher serum concentrations were likely contributed by the loading dose. Since ketamine is a high clearance drug, it requires continuous infusion for at least five half lives, which is approximately 10 h to maintain steady state [37]. Ketamine also increases sympathetic nervous system activity, resulting in increased cardiovascular functions and hepatic blood flow [38, 39]. Thus, the elevated heart rates in the present study are likely a reflection of sympathetic stimulation by ketamine.

Nociceptive mechanical thresholds increased significantly compared to baseline after administration of the K30, KL30 and K50 groups and remained elevated throughout the 120-minutes of infusion. The thresholds returned to baseline at 140 min, that is, within 20 min after the end of infusion. Mechanical thresholds have been used in previous studies in animals [14, 26–28, 40–45] for the objective quantification of pain and to evaluate the analgesic efficacy of drugs. The analgesic effect of the test drug reported by these studies was either increase in the mechanical thresholds at the site of test compared to baseline [26, 28] or concurrent rise in the mechanical thresholds equal to that produced by positive control analgesic or dose used to compare with [40, 45] or equal rise in mechanical thresholds at the site of surgery by the same drug compared for two routes of administration [40, 44]. Kukanich et al. (2005) and Kaka et al. (2015) reported increase in mechanical thresholds after administration of intravenous morphine at 1 mg/kg [26, 28]. The increase in mechanical thresholds compared to baseline in these studies was suggestive of analgesic action of morphine. In the present study mechanical thresholds increased significantly compared to baseline in K30, KL30 and K50 treatments throughout the infusion period of 120 min and returned to baseline within 20 min after the end of infusion. The rise in mechanical thresholds suggests the antinociceptive actions of K30, KL30 and K50 throughout the duration of infusion. The modified algometer and technique used in this study have been validated previously in our lab [28]. In this study, conscious vocalization may not be a good indicator of nociceptive threshold as it occurred only in 2 of subjects, and therefore we depended on algometer readings, even though vocalization has been accepted as part of the pain study protocol in human beings.

Mechanical pressure for quantification of surgical pain or hyperalgesia is termed as “algometry”. Algometry has been used to identify abnormal pain thresholds and to monitor response to analgesic treatment in animals and humans [46–49]. Algometry has also been used to evaluate the analgesic effects of drugs in dogs [26, 27, 40]. Rate of force of application plays a crucial part on the reliability of this method. Application of force at a faster rate may provoke a low false threshold reading [50]. In a study of reliability and validity of algometer Kinser et al. (2009) concluded that this is highly reliable and valid method [51]. The authors suggested that with some practice, an individual becomes reliable in applying force with an algometer. In this study researcher who operated the algometer has sufficient practice in applying the rate of force of application during preliminary experiments, which have been demonstrated as consistent results in saline treatment during validation of the method [28]. Similarly, in this study no significant difference in mechanical thresholds between control and 140 min indicates the reliability of results. Thus, the results of the above studies suggest that algometry is reliable and valid technique for the quantification of mechanical nociceptive thresholds. It should be noted that despite an open label approach of the current study, potential operator bias was minimized by blinding the operator of the algometer to the threshold values. Only another nearby observer who was not operating the algometer was entrusted to record the readings displayed on the LCD screen. It should be noted that the operator of the algometer, cannot see the values projected on the LCD screen. We are aware that open label studies have the limitation that results could be refined and biased [51], however, we tried to minimise the bias by blinding to the treatment the investigator who evaluated sedation and side effects of the infusions, and the investigator operating the algometer to the values recorded.

Ketamine has been reported to increase body temperature in dogs [23]. Effects of body temperature on mechanical thresholds in dogs have not been reported. In this study body temperatures increased significantly from baseline throughout study, however, this increase was within the normal range. The increase in mechanical thresholds throughout infusion and decrease in the mechanical thresholds after the end of infusion, at which time the temperature was still higher than baseline suggests that increase in temperatures did not affect the mechanical thresholds.

Ketamine is well known for its psychomimetic effects in dogs; these effects range from hypersalivation to excitability and delirium. The dose regimens used in this study did produce such side effects; however they were harmless and ceased within 20 min of ending the infusions. Furthermore, side effects were less intense and occurred less frequently in K30 and KL30 compared to K50. The sedation score recorded in this study was as low as 0 in K30 and KL30, and as high as 3 in K50, which is mild and comparable to the scores reported by Bergadano et al. [20].

Commonly used potent analgesics reduce the minimum alveolar concentration (MAC) of inhalant anaesthetics in dogs and humans [52, 53] ; the reduction in MAC of inhalant anaesthetic has been suggested either due to analgesic or sedative effects of the drug. Hence, if a drug decreases the MAC, it should be tested for its analgesic effects in awake patients [54]. The dosage of lidocaine selected in this study was based on MAC studies in dogs under anaesthesia [25] and studies in conscious dogs [55]. In the study on conscious dogs, lidocaine CRI of 100 μg/kg/min up to four hours did not produce any side effect [55]. Therefore, lidocaine at 2 mg/kg loading dose and CRI 100 μg/kg/min was combined to ketamine in this study. This combination of lidocaine and ketamine in present study was tested in preliminary study on two dogs before actual experiments. The major benefit of combining lidocaine and ketamine is to decrease the drug related side effects [56], decrease opioid requirement [57] intensity of pain in postoperative period due to central sensitization during surgical intervention [21]. The addition of lidocaine at 2 mg/kg loading dose and CRI of 100 μg/kg/min to ketamine infusion in this study resulted in higher mean nociceptive threshold values only at carpal pad and metacarpal footpad. Whereas, no difference in the serum concentration between K30 and KL 30 as well as KL 30 and K50 was observed due to addition of lidocaine to ketamine. The inherent high individual variability in these parameters would require large sample size to demonstrate statistical difference, which was not feasible in this study.

In this study, overall nociceptive thresholds were relatively higher in K50 than K30. Suggesting that K50 provided slightly more analgesia than K30. On the other hand, there was no such difference between KL30 and K50. Thus, K 50 and KL30 demonstrated more or less similar effect on nociceptive thresholds than K30. Nevertheless, all the three regimens significantly raised nociceptive thresholds from baseline thresholds, suggesting significant analgesic effects demonstrated by the three regimens used. Further studies are required to investigate the analgesic effects of these three doses in dogs undergoing surgical procedures.

All the three treatment regimes in this study maintained serum concentrations of ketamine above 200 ng/mL and raised mechanical nociceptive thresholds at most body points during the 120 min of infusions. A linear correlation between the mechanical thresholds and serum concentrations of ketamine was found. Results from this study suggested that the minimum concentration of ketamine to produce analgesia in dogs is between 100 and 200 ng/mL. All the three treatments provided antinociceptive effects throughout the infusions. In this setting the addition of lidocaine to ketamine increased mechanical thresholds without any further side effect. Clinical studies are encouraged to evaluate the analgesic effects of ketamine CRI and combination of lidocaine and ketamine at these dose regimens.