Although dog and sheep models have been used successfully in cardiovascular research because of their cardiovascular stability and tolerance to cardiopulmonary bypass (CPB) procedures, the anatomical dissimilarity between humans and these species and high costs limit their use in preclinical trials. Because pigs are superior to other species with respect to anatomical and physiological similarities to humans, such as heart size and surgical anatomy, cardiac output and blood pressure, they are good candidates for preclinical cardiovascular study [1–3].
Anaesthesia of the pig poses a great problem for experimental animal-based research and especially for those that need CPB procedures. On one hand, many anaesthetics have a direct effect on cardiopulmonary function. Anaesthesia-related death is most often associated with respiratory depression mediated centrally in the brain by the anaesthetics given and is followed by hypoxia and hypercarbia. On the other hand, CPB with cardiac arrest and aortic clamping induce a marked reduction of heart function in the 24 h postoperative period, and complications during and after this procedure constitute a major challenge .
Anaesthetics, such as ketamine and pentobarbital, are widely used in animal experiments. The pharmacological properties of the two agents are different. The former produces intense analgesia, sympathetic nervous system stimulation, and increased blood pressure and heart rate. When used as a monoanaesthetic, ketamine does not appear to induce surgical anaesthesia in pigs . Pentobarbital is an adjunct for anaesthesia and has no analgesic effect at normal doses. It needs to be supplemented with other analgesics to remove pain. Large doses will not achieve proper anaesthesia and could also cause the significant complications of myocardial depression and hypotension . A reasonable approach, therefore, would be to combine these two agents, utilizing their analgesic and hypnotic properties in mutual enhancement in order to reduce their concentrations at which the side effects described for the two substances do not occur. Yet, so far, the ketamine–pentobarbital combination has not been investigated for long-term anaesthesia in miniature pigs undergoing CPB procedures.
The purpose of the present study was to develop a suitable anaesthesia model in which miniature pigs could be induced smoothly and maintained stably during and after a CPB procedure. An additional goal was to propose a practice method for assessment of anaesthetic effectiveness in the miniature pig.
This study was conducted with the approval of our ethics committee. All animals received humane care in compliance with the ‘Principle of Laboratory Animal Care’ (National Institutes of Health publication no. 85-23, revised 1996).
Hybrid, female or male-castrate, healthy minipigs (body weight 25.6 ± 3.4 kg) were obtained from the Institute of Laboratory Animals, Beijing Academy of Agriculture Sciences, Beijing, China. All animals were given free access only to water, 12 h before anaesthesia.
Anaesthesia and experimental set-up
Thirty-one pigs were randomly divided into either the ketamine–pentobarbital group (K-P) or the pentobarbital group (P) before anaesthetic induction. All animals were premedicated with atropine 0.04 mg kg−1 and diazepam 0.4 mg kg−1. Then, 15 animals in group K-P were induced with intramuscular injections of ketamine 5 mg kg−1 and pentobarbital 20 mg kg−1, whereas 16 in group P were induced with pentobarbital 30 mg kg−1 alone.
After 8–15 min, the animals were transferred to the operating room table and immobilized in a supine position and monitored by continuous electrocardiogram (ECG, BASH monitor, GE Medical Systems Information Technologies, Inc., Wisconsin, USA). Then, they were endotracheally intubated with a 6.0–7.0 mm tube (Mallinckrodt Medical, Inc., Athlone, Ireland) and coupled to a ventilator (Siemens Servo Vetilator 900C, Siemens, Berlin, Germany). The skins were shaved, cleaned, and disinfected at the surgical sites. Within the ventral neck area, the skin was incised paramedially. The intravenous (i.v.) access was obtained using an 18–16 gauge double lumen venous catheter (Arrow International, Inc., Reading, Pennsylvania, USA) in the left internal jugular vein for pressure readings and fluid administration and a 20-gauge arterial catheter (Arrow International, Inc.) in the left carotid artery for arterial pressure monitoring and blood sampling.
Following venous catheter implantation, anaesthesia was maintained by continuous infusions of ketamine (3–5 mg kg−1 h−1) and pentobarbital (6–8 mg kg−1 h−1) in group K-P. However, the infusion of pentobarbital was withdrawn after CPB started. For group P, anaesthesia was maintained throughout the operations by continuous infusion of pentobarbital (8–10 mg kg−1 h−1). In both groups of animals, midazolam was injected hourly at a dose of 0.1–0.2 mg kg−1 and pipecuronium at a dose of 0.1 mg kg−1 h−1 to achieve optimal surgical conditions. The nasopharynx temperature (NPT) of the animals was continuously monitored by a thermistor probe (Marquette Electronics, Inc., Milwaukee, Wisconsin, USA), and urine output was recorded through an indwelling 12 Fr urinary catheter. Ceftazidime 1.0 g was given as prophylactic antibiotic.
A Siemens 900C ventilator was used for mechanical ventilation in the synchronized intermittent positive pressure mode with a positive end-expiratory pressure (PEEP) of 5 cmH2O and inspiratory O2 fraction (FiO2) of 50–80%. Tidal volume (Vt) was kept at 10 ml kg−1, and frequency was adjusted to maintain end-tidal CO2 pressure (PetCO2) between 35 and 45 mmHg. Arterial blood gas analysis was carried out periodically, and ventilatory adjustments were made to maintain blood gas within normal ranges.
Cardiopulmonary bypass materials and procedure
After achieving general anaesthesia, the heart was exposed by means of a standard median sternotomy and 4 mg kg−1 heparin was injected intravenously in preparation for CPB with a target activated clotting time (ACT) of at least 480 s. The pericardium was dissected free of the fat and opened in a midline, reverse-T-shaped fashion. Stay sutures were inserted. An 18 Fr arterial cannula was inserted into the ascending aorta, and a 6 Fr aortic root cannula was inserted above the aortic valve for cardioplegia. A 26 Fr venous cannula was inserted into the superior vena cava through the right atrial appendage and a 28 Fr cannula was inserted (Terumo Cardiovascular Systems, Corp., California, USA) into the inferior vena cava through a stab wound at the cavoatrial junction. Afterwards, lines were connected and CPB was instituted (Sarns roller pump; Sarns, Ann Arbor, Michigan, USA; Kewei-II hollow fiber membrane oxygenator, Kewei, Dongguan, China) with flows of 2.5–3.5 l min−1 and was adjusted according to blood gas parameters. CPB was initiated and the body temperature of the animal was decreased to 32°C. The aorta was clamped, and cold hyperkalaemic blood cardioplegia (4°C) was given immediately until the heart completely arrested. The tricuspid valve was exposed by a right atriotomy incision.
After completing most of the tricuspid valve repair, rewarming was begun. The right atrium was closed, air was evacuated, the aortic clamp was released, and the heart was defibrillated. When hemodynamic stabilization was achieved, CPB was discontinued, followed by removal of the cannulas and administration of protamine. Then, haemostasis was accomplished in a systematic fashion. The chest was closed in layers after insertion of a thoracic drain tube that was kept in place for a few hours. Blood from the CPB circuit was subsequently transfused.
Anaesthesia and assisted ventilation were continued until the animals were haemodynamically stable and had acceptable blood gas levels. Extubation was carried out when the respiratory rate was greater than 12 min−1, and end-tidal CO2 was less than 6.5 vol%. Fluids (lactated Ringer's solution) were given at the basal rate of 2–4 ml kg−1 h−1. Furosemide was given in doses of 5–10 mg to maintain an hourly urine output above 2 ml kg−1. Cardiac dysfunction [mean arterial pressure (MAP) < 60 mmHg and central venous pressure (CVP) > 10 mmHg] was treated with 6–12 μg kg−1 h−1 epinephrine. If sinus bradycardia (heart rate < 80 beats min−1) occurred, isoproterenol was used in doses of 4–8 μg kg−1 h−1 to maintain a heart rate above 100 beats min−1. Antibiotic therapy consisted of i.v. ceftazidime 1.0 g, two times daily for approximately 3–4 days.
The animals were allowed to recover and were kept under observation in our controlled animal facility, where the general health of the pigs was checked daily. Any signs of pain, infections or heart failure (fatigue, dyspnoea, coughing) were immediately reported and managed appropriately. Animals were followed for 6–9 months postoperatively.
Monitoring for anaesthesia included ECG, MAP, CVP, respiratory rate, NPT, and arterial blood samples collected for gas analysis. All these parameters were studied at various observed time points, which included 10 min after induction (T0), immediately before incision (T1), immediately before CPB started (T2), 10 min after aortic cross-clamping (T3), 10 min before the end of CPB (T4), immediately before chest closure (T5), 120 min after chest closure (T6), and 10 min before (T7) and after (T8) extubation. Afterwards, the following variables were also recorded: the time of anaesthetic induction and maintenance, aorta cross-clamping and CPB time, skin-to-skin time and chest drainage volume, and time on the ventilator as well.
Assessment of anaesthesia
Assessment of effectiveness of anaesthesia took into account respiratory frequency and intensity, cardiovascular function, ocular signs, pain stimuli, and muscle tone . The classification of anaesthetic effectiveness was defined as follows.
During induction, pigs were well tranquillized. Breathing was regular with equal contributions from chest and abdomen. The palpebral and corneal reflexes disappeared. The withdrawal reflex was absent, and the laryngeal reflex also disappeared. Jaw tone decreased greatly, and intubation was performed successfully. During the operation, surgery or deep anaesthesia was maintained with good muscular relaxation. Surgical stimuli did not produce any response, and there was no cardiopulmonary distress. During the recovery phase, pigs were completely awake and haemodynamically stable before extubation and had regained full muscle strength and spontaneous breathing with acceptable oxygenation and ventilation.
During induction, the pigs were slightly restless. Respiration was frequently irregular. Jaw tone was appreciably intense, and laryngospasm appeared during intubation. During the operation, relaxation was not very good, and the haemodynamics slightly changed with surgical stimuli. During the recovery phase, anaesthesia was not unnecessarily prolonged, but the pigs showed a little restlessness and were not haemodynamically stable.
During induction, pigs were obviously restless. Stress reactions were too intense to intubate, or respiratory and cardiac arrest happened rapidly, with no time to finish intubation. During the operation, respiration and circulation were unstable or were obviously distressed. During the recovery phase, anaesthesia was prolonged, and the peripheral circulation was poor with a falling MAP. Deep extubation had to be performed with delayed awakening, and apnoea occurred occasionally after extubation.
Continuous variables were expressed as means ± standard deviation, and categorical variables were shown as a percentage. Comparisons between groups were made with two-tailed, unpaired Student's t-test for the continuous variables and with two-tailed Fisher's exact test for the categorical variables. Analysis of variation (ANOVA) was used to compare the statistical differences between values at baseline and various time points within groups. A P value of less than 0.05 was considered statistically significant (SPSS 11.5 software, SPSS, Inc., Chicago, Illinois, USA).
General operative data
The general operative data are summarized in Table 1. In group K-P (ketamine–pentobarbital anaesthesia), the heart rate was significantly higher than in group P (pentobarbital anaesthesia) at 10 min after induction (124 ± 8 versus 115 ± 12, P < 0.05). Similarly, in group K-P, the respiratory rate was also higher at T0, T7, and T8 (14 ± 3 versus 11 ± 4, P < 0.05; 32 ± 9 versus 24 ± 8, P < 0.05; 27 ± 7 versus 20 ± 7, P < 0.05, respectively), whereas the times for induction, maintenance and on the ventilator were significantly lower than those in group P (9 ± 2 versus 12 ± 2, P < 0.001; 205 ± 33 versus 317 ± 41, P < 0.001; 372 ± 76 versus 655 ± 96, P < 0.001, respectively).
No significant differences were found in age, body weight and skin-to-skin time between the two groups. Although the aortic cross-clamping and total CPB time was also similar in both groups, the time from the aortic cross-clamping release until bypass was ended was markedly shorter in the K-P animals (38 ± 4 min) than in the P animals (44 ± 6 min), P < 0.05.
There were no significant differences at baseline (T1) between the two groups for heart rate, MAP, NPT, or CVP (Figs 1–4). Compared with baseline, the heart rate decreased significantly at T4 and T5 in group K-P and also decreased markedly at T4 in group P. However, the heart rate decreased more in group P than in group K-P at T4 and T6 (13 ± 3% versus 8 ± 2% and 10 ± 3% versus 5 ± 1%, P < 0.05, respectively; Fig. 1).
Although MAP decreased markedly over time in both groups compared with baseline values, it decreased more in group P than in group K-P at T4 (25 ± 4% versus 17 ± 3%, P < 0.05), T5 (21 ± 4% versus 13 ± 3%, P < 0.05), and T6 (15 ± 3% versus 9 ± 2%, P < 0.05; Fig. 2). Animals in group P were more likely to require continuous infusions of epinephrine and isoproterenol than those in group K-P.
NPT decreased at T3, recovered at T5, and increased markedly at T6 compared with baseline in both groups. In addition, it was significantly higher at T4 and T5 in group K-P than in group P (Fig. 3, P < 0.05).
No statistical differences were found between the two groups for CVP throughout the period from baseline to the subsequent time course of the experiment (Fig. 4). In group K-P, however, CVP seemed more stable than in group P.
Arterial blood gas analysis
Table 2 shows the intraoperative and early-recovery arterial blood gas parameters in group K-P versus group P. The pH values were similar in both groups at T2, T3, and T4. Thereafter, the levels were markedly different, averaging 7.47 ± 0.03 in group K-P and 7.42 ± 0.04 in group P immediately before chest closure (T5) (P < 0.05) and averaging 7.48 ± 0.03 in group K-P and 7.45 ± 0.04 in group P 120 min after chest closure (T6) (P < 0.05).
Similarly, the PaCO2 values were stable at the end of CPB. Afterwards, the levels were markedly higher at T5 in group K-P than in group P (38.3 ± 3.5 versus 35.2 ± 2.3, P < 0.05). However, the levels were significantly lower at T6 in group K-P than in group P (33.5 ± 3.0 versus 37.2 ± 4.0, P < 0.05).
The PaCO2 and base excess values were similar at T5, or during the previous time course of the experiment. Then, the two values were found to be markedly lower at T6 in group P than in group K-P.
The Glu and Lac values were kept in close agreement at T2 and T3 between both groups. Thereafter, the Glu levels were significantly elevated in group K-P compared with group P at T4, T5, and T6. Also, the Lac values were markedly lower at T4 and T6 in group K-P than in group P. The values of SaO2, K+, Ca2+, and haematocrit were in close agreement between the groups over the time course.
Anaesthesia-related complications and assessment of anaesthesia
For different types of anaesthesia, pentobarbital had greater anaesthesia-related mortality (P < 0.05), as illustrated in Table 3. All animals in group K-P were stable during the surgical procedures and survived to at least the first postoperative day. Three animals in group P died from respiratory and cardiac arrest induced by anaesthetic induction, and two died from postoperative apnoea after extubation.
With respect to anaesthesia-related complications, the morbidities of tracheotomy and reintubation were higher in group P than in group K-P, but there were no significant differences between the groups. Respiratory and cardiac arrest occurred in only one (6.7%) animal in group K-P, but in seven (43.8%) animals in group P (P < 0.05).
According to the assessment methods designed in the present study, 73.3% of animals in group K-P were evaluated as having class I (best) effectiveness of anaesthetic induction, whereas only 31.3% in group P had class I (P < 0.05). Furthermore, 80% of pigs in group K-P were assessed as having class I effectiveness of anaesthetic maintenance, in contrast to only 30.8% in group P (P < 0.05).
The ideal anaesthetic agent would provide full anaesthesia for cardiac surgery: analgesia, deep hypnosis, low cardiovascular stress, and muscular relaxation. Most injectable anaesthetics do not possess all four properties and must be used in combination to achieve full anaesthesia. Ketamine and pentobarbital are extensively used in experimental pigs. The toxicity of pentobarbital is dose dependent. Particularly negative effects of pentobarbital are cardiac and respiratory depression perioperatively and a very long recovery time postoperatively [8,9]. However, ketamine will usually stimulate rather than depress the circulatory system .
In this study, the effects of ketamine–pentobarbital anaesthesia and pentobarbital anaesthesia in minipigs undergoing cardiac surgery with CPB were compared. To our knowledge, this is the first study of its kind that has been conducted on minipigs undergoing cardiac surgery. The findings revealed significant differences between the groups during the perioperative period, from anaesthetic induction through CPB and to 24 h postoperatively.
In a previous study, minipigs were sedated with pentobarbital (10 mg kg−1, intramuscularly) and ketamine (20 mg kg−1, intramuscularly) and underwent intratracheal intubation. Then, animals were maintained by bolus injections of ketamine (20 mg kg−1 h−1) and supplemented with boluses of pentobarbital (2.5 mg kg−1 h−1). At these concentrations, no significant haemodynamic changes were demonstrated during the control anaesthesia-only period without any surgical procedure . However, the surgical interventions, such as laparotomy and intestinal manipulation, caused marked changes in heart rate and MAP. This indicated that such dosages, especially of pentobarbital, could not ensure adequate depth of anaesthesia for sophisticated procedures. In another study on the cardiovascular effects of pentobarbital in pigs, a surgical anaesthesia was performed by continuous pentobarbital infusion of 12 mg kg−1 h−1. After the dose was raised to 30 mg kg−1 h−1, cardiovascular function was impaired after 90 min infusion time .
In the present study, minipigs in group K-P were induced with low-dose ketamine (5 mg kg−1) and moderate-dose pentobarbital (20 mg kg−1). After endotracheal intubation and i.v. catheterization, the animals were maintained by continuous infusion of low-dose ketamine (3–5 mg kg−1 h−1) and low-dose pentobarbital (6–8 mg kg−1 h−1). By using the combination, ketamine and pentobarbital will produce balanced anaesthesia with synergistic effects. The two agents will mutually enhance their sedative and analgesic properties with a significant reduction in the dosages needed compared with either one alone. Thus, the side effects of these anaesthetics, such as cardiovascular and respiratory depression, could be avoided by continuous low-dose injection. Additionally, the infusion of pentobarbital was withdrawn after CPB started in the animals anaesthetized with ketamine–pentobarbital, which could shorten the recovery time postoperatively.
In this study, the general operative data confirmed the above hypothesis. The animals in the ketamine–pentobarbital group had a more stable heart and respiratory rate, faster onset time of induction, and shorter recovery time postoperatively than those in the pentobarbital group. Furthermore, the time from aortic cross-clamping release until bypass was ended was markedly shorter in the combination anaesthesia group than in the pentobarbital-only group, which meant that the pentobarbital anaesthetic pigs required a longer time on bypass circulation than the ketamine–pentobarbital anaesthetic ones before stable haemodynamics was obtained.
It has been suggested by many studies that anaesthesia is not the only mechanism that would explain the cardiovascular depression in pigs. The prolonged CPB with cardiac arrest and aorta clamping also produce a detrimental effect on cardiac function. Ischaemia–reperfusion injury could be responsible for cardiac dysfunction following CPB . Repeated ischaemic insults provoke considerable myocardial oedema associated with a reduction in diastolic compliance and a significant selective decrease in subendocardial perfusion. Reperfusion injury of the heart is also manifested by reversible myocardial dysfunction (myocardial stunning), arrhythmias, and accelerated cell death [12,13].
In the present study, the haemodynamic parameters in the ketamine–pentobarbital anaesthetic animals were markedly different from those in the pentobarbital group. However, the most significant differences in haemodynamics always occurred during the time from aortic cross-clamping release until bypass was ended in both groups. Combining the above analyses, whereas anaesthesia induced a moderate reduction in cardiovascular function, CPB with aortic clamping and cardiac arrest also evoked a certain degree of reduced cardiac function.
Pentobarbital and ketamine significantly increased glucose levels in the blood within 10 min of injection . In this study, the blood glucose levels increased in both groups and were significantly higher after CPB rewarming in the ketamine–pentobarbital group than in the pentobarbital group. This difference may be related to their synergistic effects and body temperature. Additionally, the high-dose glucose injection was postoperatively contraindicated in all the pigs, because it would markedly prolong the respiratory recovery of animals anaesthetized by pentobarbital.
It is well known that an elevated lactate level represents inadequate tissue perfusion and oxygen delivery. In addition, MAP is the most accurately measured and may be most useful for assessing the perfusion pressure of vital organs. In the present study, blood lactic acid levels were significantly increased from those when the patient was on CPB until 2 h after surgery in the pentobarbital group compared with the ketamine–pentobarbital group, and the MAP was markedly lower at the same experimental time points in the pentobarbital group compared with the ketamine–pentobarbital group. These results indicate that pentobarbital anaesthesia produced more insufficiency of tissue perfusion than ketamine–pentobarbital anaesthesia in the minipigs for the CPB procedure.
At present, the depth of porcine anaesthesia is difficult to measure accurately as no single method is reliable for all anaesthetic agents. A substantial and special method still needs to be established for assessing anaesthetic effectiveness in laboratory animals. In spite of the lack of advanced monitoring equipment, we made an evaluation based on five organ systems – respiratory, cardiovascular, eye-related, mucocutaneous, and muscular – and designed a practical and simple method of assessment of anaesthetic effectiveness. In this study, animals anaesthetized by pentobarbital alone had greater mortality and morbidity from anaesthesia-related respiratory and cardiac arrest than those anaesthetized by combination anaesthesia. Regarding the classification of anaesthetic effectiveness, the percentage of animals that achieved best anaesthetic effects was significantly higher in the combination anaesthesia group than in the pentobarbital alone group.
In conclusion, minipigs could be an excellent research model in the study of cardiovascular surgery under CPB with cardiac arrest similar to the situation in humans. Although CPB with aortic clamping and cardiac arrest induced a certain reduction in cardiac function, combination anaesthesia with low-dose ketamine and pentobarbital demonstrated superior haemodynamic and respiratory indices in comparison with pentobarbital. However, further investigations are needed to compare the differences of effects between the combination anaesthesia and CPB with ischaemia–reperfusion injury on cardiac function.
The study was supported by grant no. 30772152 from the National Natural Science Foundation of China.