Determination of Minimum Infusion Rate and Cardiorespiratory Effects of Total Intravenous Anesthesia of Ketofol With or Without Lidocaine, Fentanyl or Dexmedetomidine in Dogs

Background: Ketofol is a 1:1 mixture of ketamine and propofol that has been proposed for induction and maintenance of anesthesia aiming to provide more cardiovascular stability and less undesirable impacts compared to the use of propofol and ketamine alone. However, it has been associated with exacerbated respiratory depression in dogs. Diminishing the dose of ketofol may improve cardiovascular effects and attenuate respiratory depression. The present study was designed to evaluate the effect of adding lidocaine, fentanyl or dexmedetomidine at the required dose and cardiorespiratory variables in dogs undergoing total intravenous anesthesia (TIVA) with ketofol. In phase I, twelve dogs were induced and maintained with two out of four anesthetic regimens of KET: ketofol (4 mg/kg and 0.3 mg/kg/min, respectively), KLD; ketofol and lidocaine (1.5 mg/kg and 0.25 mg/kg/min, respectively), KFN: ketofol and fentanyl (LD: 5 µg/kg and 0.1 µg/kg/min, respectively) and KDX: ketofol and dexmedetomidine (2 µg/kg and 2 mg/kg/h, respectively). Minimum infusion rate (MIR) of ketofol was determined in this phase. Subsequently, in phase II, other twelve dogs were given the same anesthetic regimens for 60 min similar to the previous phase, except the infusion rate of ketofol. Cardiorespiratory variables were recorded in predetermined interval. Results: decreasing baseline during


Background
Inhalation anesthesia requires some specialized equipment to deliver oxygen and remove carbon dioxide gas. Total intravenous anesthesia (TIVA) does not require such facilities, so it can be considered as a substitute for inhalation anesthesia, particularly in short-duration anesthesia as well as in minimally invasive procedures or diagnostic imaging [1]. This technique is going to be more popular, especially with new generations of anesthetic agents that having short acting and non-cumulative properties and with the advent of modern technologies providing facilities for having drugs with more e cient constant rate infusion (CRI) [2].
Propofol is the most injectable anesthetic used for the induction and maintenance of anesthesia. This drug has rapid onset, short duration and slow recovery; however, it was shown that it can be associated with dose-dependent respiratory depression and hypotension [3][4][5][6]. Ketamine is another common anesthetic drug used for small animals. Anesthesia accompanied by ketamine result in an increase in heart rate, blood pressure and cardiac output. It may also lead to muscle stiffness, convulsion and eventful recovery [1].
Combination of propofol and ketamin has been proposed to be used for TIVA diminishing the doses required for each drug and subsequently providing more stable cardiorespiratory variables along with less unpleasant consequences. Notably, Ketofol is a combination of 1:1 propofol and ketamine in a single syringe. Human studies revealed that ketofol could increased hemodynamic stability [7] and decrease complications rates associated either with propofol or with ketamine [8,9]. In a study performed on dogs, TIVA with ketofol resulted in higher heart rate (HR) and acceptable mean arterial pressure (MAP) compared to the administration of propofol alone; meanwhile, the respiratory depression was more pronounced in ketofol treatment than propofol [1]. Respiratory depression has also been observed in those dogs induced either with ketofol alone or in combination with diazepam or midazolam [10,11].
Various drugs have been used in combination to anesthetic agents to provide a balanced anesthesia with the bene t of the reduced dose of the used drug(s), and subsequently, to avoid unpleasant effects associated with each one. Lidocaine is an amino amide local anesthetic reducing the amount of inhalant anesthetic drugs with no signi cant cardiovascular impacts [12,13]. The infusion of lidocaine with ketamine in dogs undergoing TIVA with propofol has diminished the dose rate of propofol [2]. Fentanyl is a pure µ opioid agonist that has rapid and short-lived impacts. It was indicated that fentanyl has reduced the minimum alveolar concentration of volatile anesthetics [14][15][16] and the induction dose and minimum infusion rate (MIR) of propofol and alfaxalone in the dogs [17,18]. Dexmedetomidine (DEX) is a potent alpha-2 adrenergic agonist used due to its sedative and analgesics effects on dogs [19]. Moreover, DEX has been shown to reduce the dose of anesthetic agents needed for the induction and maintenance of anesthesia [20][21][22].
Since no published study on evaluating the effect of CRI of ketofol with either lidocaine, fentanyl or DEX on dogs was found, the present study was designed to determine the MIR and the accompanied cardiovascular alterations in dogs receiving CRI of ketofol combined with lidocaine, fentanyl or DEX. In this regard, we hypothesized that addition of either employed drugs to ketofol, via signi cant reduction in the dose rate of ketofol, would result in the improved cardiovascular variables and respiratory function.

Methods
The present investigation was designed as a blind randomized (https://www.randomizer.org) prospective experimental study. Twenty-four mongrel male dogs weighing 19.1 ± 3.1 kg and aged 1.5-2.5 years old, belonged to a private shelter designed for the protection of stray dogs, were selected and then transferred to the Veterinary Hospital at least one week before the initiation of the study. The animals were kept in separate cages, fed with the same diet, and had free access to water. The dog's health status was con rmed by a thorough physical examination, complete blood count and total protein measurements. If any dog was considered not to have status I, according to American Society of Anesthesiologists physical status, would have been excluded from the study. Other exclusion criteria were pregnancy, aggressive behavior and extremely fat or slim. Thereafter, the dogs were fasted for 12 hours, but they had free access to water for up to 2 hours before the anesthesia session. The Ethical Committee of our university has approved all the procedures performed in the current study.
The present study was conducted in two phases: Phase I-Determination of MIR of ketofol Twelve dogs were studied in this phase. On the day of the experiment, the animals were given a combination of acepromazine (0.025 mg/kg; Neurotranq, Alfasan, Woerden, Holland) and morphine (0.25 mg/kg; Morphine Sulfate, Daru Pakhsh, Iran) intramuscularly. Thirty minutes later, the animals were transferred to a surgical table and the cephalic veins of both forelimbs (for drug and serum administration) and the dorsal metatarsal artery of the left hindlimb (for blood samples collection) were catheterized using an appropriate catheters' size. The animals received 100% oxygen (2 L/min) with a mask for 5 minutes. By passing 40 minutes from sedatives injection, the dogs received either of the following anesthesia regimens: 1-KET: induction by bolus injection of ketofol (4 mg/kg) followed by an IV bolus of 2 mL normal saline and the maintenance by CRI of ketofol (0.3 mg/kg/min) and normal saline (0.0125 mL/kg/min), 2-KLD: induction by bolus injection of ketofol (4 mg/kg) followed by an IV bolus of lidocaine (loading dose (LD); 1.5 mg/kg ;Lignodic 2%, Caspian Tamin, Iran) and the maintenance by CRI of ketofol (0.3 mg/kg/min) and lidocaine (0.25 mg/kg/min).
Each dog received two out of four treatments with at least one-week interval. The preparation process of ketofol was performed according to the study by Kennedy and Smith [1]. Ketofol and the accompanied drugs' infusions were performed using a syringe infusion device (Medifusion, DS-3000, South Korea) and a burette infusion microset (60 drops:1 mL; Shanchuan medical instrument Co. China), respectively. Afterwards, LD of lidocaine, fentanyl and DEX were diluted to the nal volume of 2 mL using normal saline.
After the induction of anesthesia, the trachea was intubated using a proper-sized cuffed endotracheal tube, which was then connected to a rebreathing anesthetic system delivering oxygen at 2 L/min. The dogs were positioned in sternal recumbency and allowed to breathe spontaneously. All the animals received normal saline at the dose rate of 10 mL/kg/h from the induction of anesthesia until the complete removal of the tracheal tube. It was attempted to maintain dogs' rectal temperature at 37-38 °C from anesthesia induction until the removal of the tracheal tube using a blanket and two warm water bags that were placed over and at both sides of the animals, respectively.
For determining MIR of ketofol, a noxious stimulus was applied between the ngers of the animal using a hemostat forceps closed at the rst ratchet for 2-3 seconds. If the dog had no purposeful movement, the infusion rate was reduced by 0.05 mL/kg/min. In contrast, if the dog had any purposeful movement, the infusion rate was increased by 0.05 mL/kg/min. The dog remained at the new rate for 10 minutes and then painful stimulation was applied once again. This process continued until there were two stimuli that caused and did not cause purposeful movements. MIR was calculated based on the arithmetic average of the two above-mentioned rates. Purposeful movement is de ned as any movement leading to an obvious movement of the head and trunk as well as limb's withdrawal. Tremor and twitching were not considered as purposeful movements.
Arterial blood sampling was done by passing at 40 minutes from sedation (before the induction of anesthesia), 5 and 15 minutes after the induction of anesthesia, and then every 15 minutes until obtaining MIR. In this phase, no data were recorded except infusion rate were recorded. The blood samples were discarded with no further analysis. Ketoprofen (2 mg/kg) was postoperatively administered to all the dogs q24h for a 2-day.

Phase 2-Determination of anesthesia scores, recovery times and cardiorespiratory effects
Twelve other dogs were used in this phase. Sedation, instrumentation, dogs' monitoring and ketofol preparation were done as described earlier for phase 1. Thereafter, dogs were then randomly assigned either to receive the two treatments of KET, KLD, KFN, or KDX with one week for washout. The rate of ketofol was selected based on the results obtained from the rst phase. Anesthesia was maintained for 60 min and then the infusion was disrupted, but oxygen supplementation was continued until extubation.
The removal of tracheal tube was done once swallowing re ex returned or tongue movement was seen.
Bradycardia (HR<40 beats/min) was treated by atropine (0.04 mg/kg) as needed. If apnea was seen for more than 5 minutes or ETCO2 was >55 mmHg, intermittent positive pressure ventilation (IPPV) would be performed to restore ETCO2 within the normal range (i.e. 35-45 mmHg).
In this phase, sedation, anesthesia induction, and recovery qualities were scored in terms of the following scoring systems: sedation quality: 1-no sedation, 2-mild sedation, 3-moderate sedation, and 4-heavy sedation; induction quality: 1-no sign of excitement, intubation within 60 seconds after the administration of ketofol; 2-mild signs of excitement, some struggle during intubation; and 3hyperkinesia, restlessness, no intubation; and recovery quality: 1-no struggling, standing and walking with no di culty; 2-some struggling, long-lasting sternal recumbency; 3-sever struggling, inability to be in sternal recumbency or to walk. The times for tracheal intubation, the removal of the tracheal tube, uprising the head, sternal recumbency and ability to stand (at least for 10 seconds without assistance) were recorded. HR (via SPO 2 probe connected to the tongue), blood pressure (invasive arterial blood pressure using a manometer attached to the catheter of dorsal metatarsal artery), respiratory rate (via a capnograph, f R ), and rectal temperature (via a probe connected to the mucus of rectum with the same insertion length in all the included dogs, RT) were recorded. HR, lead II electrocardiogram (EKG), f R and RT were recorded using a multiparameter monitoring system (Vitapia 7000KV, Trismed, South Korea). Correspondingly, the above-mentioned variables were continuously measured but they were recorded at 40 minutes after sedative administration (before induction of anesthesia), and also at 5, 15, 30, 45 and 60 minutes after the induction of anesthesia.
For blood gas analysis, a 1 mL arterial blood sample was taken via the dorsal metatarsal artery catheter at each time point similar to hemodynamic variables except for before the induction. For blood collection, at rst, a 1 mL blood was withdrawn in a syringe, then a 1 mL additional blood was taken in another 1-mL syringe containing 0.1 mL heparin (Heparodic 5000, Caspian Tamin, Iran), and then the rst collected blood plus 1 mL normal saline were slowly injected into the artery. Blood gas analysis was performed using a calibrated gas analyzer (EDAN i15, China).
All the drug' calculations and injections were done by one investigator (H.I.R). Two other investigators (M.K and M.K) who were unaware of the treatments have scored sedation, intubation and recovery, and also measured and recorded data. Ketoprofen (2 mg/kg) was postoperatively administered to all the dogs q24h for a 3-days period.

Statistical analysis
All statistical analyses were performed using SPSS software version 24 (IBM Corporation, USA). Normal distribution of data was assessed using Kolmogorov-Smirnov test. For normally distributed data, repeated measure for ANOVA (followed by Bonferroni's teat as needed) was used to compare the changes between and within the treatments. Friedman test was also used to compare the data on sedation quality, induction and recovery between the treatments. Parametric and non-parametric data were expressed as mean (standard deviation (SD)) and median (maximum=minimum), respectively. The statistically signi cance level of the data was considered as p <0.05. The power of the study by considering α of 0.05 was greater than 80% for all variables, which was determined by G*Power software version 3.1.9.7 (Universität Kiel, Germany).

Phase 1
There was no signi cant difference between the treatments with respect to body weights (17.3 ± 2.7 kg, 17.1 ± 2.9 kg, 19.4 ± 3.5 kg and 17.3 ± 2.2 kg in KET and KLD, KFN and KDX treatments, respectively; p>0.05). In case of MIR, the addition of lidocaine, fentanyl or DEX to ketofol reduced the MIR of ketofol by 37%, 50% and 80%, respectively. The MIR of ketofol was 0.35 ± 0.04 mg/kg/min for KET, 0.23 ± 0.03 mg/kg/min for KLD, 0.15 ± 0.04 mg/kg/min for KFN and 0.08 + 0.02 mg/kg/min for KDX (p=0.001). The dose rates of 0.35, 0.23, 0.15 and 0.06 mg/kg/min were selected as the infusion rates of ketofol for KET, KLD, KFN and KDX for phase 2, respectively. The scores for sedation, induction and recovery qualities are summarized in Table 1. Overall, sedation, induction and recovery were acceptable in all the dogs and no noticeable complication was detected in any treatment. Although the induction and recovery scores gained higher quality in more dogs in KFN compared to KDX, no signi cant differences were detected regarding to evaluated qualities in the treatments (p>0.05). Table 2 shows the determined events in the recovery time. The times of head uprising and sternal recumbency were signi cantly shorter in KFN compared to those of KLD (p=0.038 and 0.003, respectively). Besides, the time of ability to stand was signi cantly lower in KFN and KDX than KLD (p=0.016 and 0.001, respectively). Induction Recovery Second and third degree atrioventricular heart blocks were observed in KFN and KDX in several cases.

Discussion
The dose of ketofol to induce anesthesia (i.e. 4 mg/kg) was selected as the same dose used in the study by Kennedy and Smith [1] of TIVA with ketofol in dogs. The dose of ketofol for the maintenance of anesthesia in the aforementioned study was 0.3 ± 0.1 mg/kg/ min, which is comparable to the dose of ketofol in the KET treatment (i.e. 0.35 mg/kg/min) in the present study. The doses of lidocaine, fentanyl and DEX were chosen based on some previous investigations performed on dogs [2,18,23,24].
In the present study, the addition of lidocaine to ketofol diminished the MIR of ketofol by 37% (0.35 ± 0.04 mg/kg/min for KET vs. 0.23 ± 0.03 mg/kg/min for KLD). Mannarino et al. [2] reported that the addition of intravenous infusion of lidocaine led to a decrease of 18% in MIR of propofol in dogs: nevertheless, this difference was not statistically signi cant. However, in the aforementioned study, concomitant infusion of ketamine and lidocaine to propofol reduced the MIR of propofol by 37%. It has also been shown that co-administration of lidocaine to animals undergoing inhalation anesthesia could reduce the minimum alveolar concentration by 10 to 45% [12,13,25,26]. The mechanism by which lidocaine reduce the amount of anesthetic agents has not been well elucidated yet. Of note, the ability of lidocaine to reduce the magnitude of anesthetics agents has been attributed to both sedative [27] and analgesic properties [28] of lidocaine.
Fentanyl decreased the MIR of ketofol by 50%. This decline was expectable in MIR of ketofol; however, the magnitude of this decrease is of more importance. There are several reports in dogs indicating the effect of fentanyl on decreasing MAC of volatile anesthetics (in a range of 35-66%) [14][15][16]. It has been proposed that co-administration of propofol and fentanyl would be more potent in the reduction of the amount of anesthetic agents than that of the combination use of an inhaled anesthetic and fentanyl. In addition, fentanyl has diminished propofol requirement for the prevention of movement by about 51 and 63% following the application of low (0.1 µg/kg/min) and high (0.2 µg/kg/min) doses, respectively [18].
By comparing fentanyl and lidocaine, it was appeared that fentanyl is able to decrease the MIR of ketofol more than that of lidocaine. This nding is in accordance with the results of the study by Steagal et al. [14] who reported a greater sparing effect of fentanyl on dogs anesthetized with iso urane than lidocaine.
In the current study, DEX diminished the MIR of ketofol more enormously (i.e. 80%) compared to lidocaine (i.e. 37%) and fentanyl (i.e. 50%). DEX has been found to decrease MAC of halothane and iso urane in dogs up to 90% and 60% dose-dependently, respectively [21,22]. Two other studies have also reported a reduction in MAC of iso urane by 86% [29] and 89% [30]. DEX appears to be superior to lidocaine and fentanyl in terms of MIR reduction of ketofol. Consistent with the results of the current study, in a study conducted on pigs, the combination use of ketamine-propofol-DEX has been observed to have more analgesic properties compared to the time when fentanyl was substituted to DEX [31].
The evaluation of induction and recovery qualities showed no signi cant differences among the treatments; however, the median recovery score were lower in the KET and KLD than those of KFN and KDX. The recovery events generally took longer times to occur in KLD treatment. Mannarino et al. [2] have also reported a prolonged recovery period in dogs anesthetized with propofol and lidocaine compared to propofol alone. The prolonged elapsed time to extubation and standing have also been reported in dogs anesthetized with sevo urane and high doses of lidocaine [12]. Although the cause of this observation was not clearly investigated in dogs, lidocaine-induced sedation as a reason for the prolonged recovery and lidocaine-induced muscle relaxation as a cause of poor quality of recovery have been proposed in horses (Valverde et al., 2005). A similar mechanism might also play a role in the dogs. In case of lower recovery times in KFN and KDX, lower infusion rate of ketofol seems attributable as the main reason; however, despite a signi cant reduction in the rate of propofol infusion, Davis et al. [18], found no difference with respect to the time to walk between propofol and propofol-fentanyl anesthesia in dogs.
In the present experiment, heart rate signi cantly increased in both KET and KLD treatments over time compared to baseline. Furthermore, in both treatments, MAP was relatively stable during anesthesia compared to baseline. Accordingly, these results are consistent with those of previous studies in which co-administration of ketamine and propofol or using ketofol lead to the increased heart rate and improved blood pressure in dogs compared to the administration of propofol alone [1,32,33]. The observed result can be attributed to the cardiovascular effects of ketamine, which results in higher HR and MAP through catecholamine release [34]. However, the addition of lidocaine or lidocaine-ketamine to dogs under propofol anesthesia did not alter the effects of propofol on HR and MAP [2].
HR and MAP in KFN treatment remained relatively stable during anesthesia, which were near to the baseline values. Even though propofol has been proposed to decrease HR dose-dependently via vagal stimulation [35] and to provide stable MAP [36], we did not expect such a great impact of fentanyl on ketofol when compared to the results obtained from the KET and KLD treatments. It seems that fentanyl has blunted the potential bene t of ketamine on HR in KFN. The same scenario has just happened for KDX. Those dogs treated with ketofol and DEX showed a mild decrease in HR and an increase in MAP during anesthesia session compared to baseline. DEX has shown to be associated with bradycardia and hypertension in dogs through parasympathetic activity and vasoconstriction property, respectively [37].
Again, we predicted that ketamine is able to prevent the decrease in HR following the DEX administration, but it appears that DEX effects were predominant over ketamine. The results regarding HR in KFN and KLD is highly likely due to the tremendous decrease of MIR of ketofol in these treatments, which subsequently resulted in a great decrease in the magnitude of ketamine to a level that was not effective enough to impact cardiovascular alterations signi cantly.
In the present study, the respiratory rate was non-signi cantly lower in KET and KLD and signi cantly lower in KFN and KDX during anesthesia compared to baseline. Additionally, there was a signi cant decrease in pH and a signi cant increase in PCO2 during anesthesia compared to the baseline values in these four treatments. The decreased respiratory rate, the decreased pH and the increased PCO2 are known as the indicators of respiratory depression, which has occurred in all the treatments. The depressant effect of the propofol and ketamine combination on respiratory system has been documented in previous studies [1,33,38]. It seems that ketamine and propofol when used concurrently has an additive depressant effect on respiration [1]. Propofol decreases the ventilatory response to CO2 and arterial hypoxemia by affecting central chemoreceptors [39]. On the other side, ketamine also affects the ventilatory response to CO2 depending on the concentration of the drug in the central nervous system [40]. In addition, ketamine μ opioid receptors' effects can also lead to respiratory depression [41].
Respiratory depression did not decrease despite ketofol dose reduction in the KLD, KFN and KDX treatments. Besides, it seems that respiratory depression tended to be more severe in those dogs receiving lidocaine. Intravenous infusion of lidocaine in the dogs anesthetized with iso urane did not change the respiratory rate and the other related parameters [14,42]. The reason for lidocaine depressant effect on ketofol is still unclear for the authors and needs to be more investigated by further studies.
Respiratory depression was also detected in KFN and KDX. Respiratory depression has been reported in the dogs receiving fentanyl and iso urane anesthesia [43,44]. Opioids has been shown to affect hypoxia and hypercapnia responses in the brainstem resulting in observing the prevalent respiratory depression with almost all opioids [45]. For DEX, similar to other alpha-2 agonists, although it may be considered as free of respiratory effects, its combination with other sedative and anesthetic agents has been proposed to result in respiratory depression [46], as seen in the current study.

Conclusions
The addition of lidocaine, fentanyl or DEX in the used doses' rates reduced the MIR of TIVA with ketofol in the dogs. The magnitude of this reduction was the greatest for DEX and the lowest for lidocaine. Those dogs that received lidocaine showed better preservation on HR than those that were administered with fentanyl and DEX. Notably, all three added drugs resulted in respiratory depression during anesthesia. A reduction in the dose of ketofol even as much higher as for DEX, was not able to attenuate respiratory depression induced by ketofol and because of tremendous reduction in ketamine's dose, it might be associated with some unpleasant cardiovascular consequences.