Association of midazolam with ketamine in giant Amazon river turtles Podocnemis expansa breed in captivity

PURPOSE: Evaluate the effects of two anesthetic associations in giant Amazon river turtles (P. expansa). METHODS: Twenty P. expansa, healthy, of both sexes, with weights between 1.0 and 1.5 kg of a commercial breeding facility located in the valley of the Araguaia River, Goiás, Brazil, were divided into two groups ( G1 n = 10 and G2 n = 10). Each group received a protocol being: P1 = midazolam (2 mg/kg IM) and ketamine (20 mg/kg IM) and P2 = midazolam (2 mg/kg IM) and ketamine (60 mg/kg IM), applied on G1 and G2, respectively. The drugs were applied in the left forelimb. The clinical parameters evaluated were: locomotion, muscle relaxation, response to pain stimuli in the right thoracic and pelvic members and heart rate. These assessments were made at time 0 (immediately after injection) and times of 5, 10, 20, 30, 45, 60, 90, 120, 150 and 180 minutes after the injections. RESULTS: Group 2 showed a higher heart rate than G1 and more rapid and prolonged immobilization. CONCLUSION: The sedation scores obtained by these protocols (P1 and P2) were satisfactory, with possible pharmacological contention for collecting biological samples and physical examination in P. expansa.


Introduction
Bennett 1 and Boyer 2 when studying anesthesia in Testudines, noted that many drugs have been tested, but most showed limitations and side effects.According to Bennett 1 , anesthesia of reptiles is still an imprecise science, compared with birds and mammals, because the results tend to be highly variable, mainly because they are heterothermics.Additionally, this author states that anesthesia of reptiles is often complicated by the long induction and recovery periods and the difficulty of monitoring parameters.Some works on anesthesia, in P expansa species, have been performed by brazilian researchers.Santos et al. 3 evaluated the effects of the combination of xylazine with propofol and Carvalho and Santos 4 developed associations with tiletamine and zolazepam and ketamine with xylazine.
Dissociative anesthetics and benzodiazepines are some of the most used drugs for anesthesia and chemical restraint in reptiles 5 .
The dissociative are used in high doses to achieve anesthesia and have long recovery periods.Already benzodiazepines are used mainly in association with the derivatives of phencyclidine 5,6 .
The midazolam is a water soluble drug, anxiolytic and anticonvulsant.Compared to diazepam it is 3-4 times more potent, and present faster metabolism.It has effects such as hypnosis, sedation, amnesia, muscle relaxation, low toxicity and does not alter heart rate, and cannot be used alone for anesthesia 7 .
The ketamine and the tiletamine, only representatives from the class of phencyclidine, are currently in clinical use.
The widespread use of these agents in veterinary medicine is due to the fact that they have high safety margin when compared to barbiturates, because there is the possibility of administration by other routes, in addition to intravenous, and for being used in several home and wild animal species 6 .
According to Mello and Cordeiro 8 , ketamine increases heart rate, causes severe analgesia in the skeletal muscle system, but does not cause muscle relaxation nor loss of protective reflexes, keeping eyes open and pupils fixed.
The use of associations of ketamine with midazolam or ketamine with diazepam have been widespread, since they cause minimal side effects 9 or no significant side effects on the cardiorespiratory system 10 .Bienzle and Boyd 7 observed that the isolated use of midazolam and ketamine in snapping turtles (Chelydra serpentina), did not produce satisfactory sedation, but when combined, these drugs are very effective in reptiles.
The objective of this paper was to evaluate the effects of two anesthetic associations, midazolam 2 mg/kg IM with ketamine 20 mg/kg IM and midazolam 2 mg/kg IM with ketamine 60 mg/ kg IM in P. expansa.

Methods
The experiment was conducted in a commercial breeding facility (15°04'18"S and 50°25'2.4"W-340 m altitude), located in the valley of the Araguaia River, Goias, Brazil.Ambient temperature was measured by maximum and minimum thermometer 1 .
Twenty specimens of P. expansa, healthy, aged approximately three years, of both sexes and body weight ranged from 1.0 to 1.5 kg, were seized from their fattening tanks with the aid of nets, weighed and individually identified.Then, the terrapins were divided into two groups (G1 and G2), each group containing 10 subjects, fasted for 24 hours before being subjected to anesthetic protocols.
Anesthetics midazolam 2 and ketamine 3  Already to obtain the heart rate it was measured the quantity of heart beats per minute, using vascular Doppler ultrasound 4 , at 0, 10, 30, 60, 120 and 180 minutes after drug administration.
All procedures and assessments of the turtles were performed with average ambient temperature of 32.7 º C (± 1.8).
The results were statistically analyzed by nonparametric Mann-Whitney U test, and the test was applied with a significance level of 5%.

Results
G1 and G2 showed no significant difference for the parameters muscle relaxation and pain sensitivity in the thoracic and pelvic, in different times (p>0.05).However, for the parameters locomotion and heart rate, differences were observed (p<0.05) at times 5, 10, 90, 120, 150 and 180 and at 10, 30, 60, 120 and 180 minutes, respectively, as shown in Figures 1 and 2.
All P. expansa at G1 reached score two (2) twenty minutes after application for muscle relaxation, remaining only 40 minutes with this score, but for the test of pain sensitivity, the anesthetic protocol P1 was unable to neutralize the pain, having response to withdrawal of members of the right thoracic and pelvic limbs to painful stimuli.
In G2 only three individuals (30%) reached score three

Discussion
Ambient temperatures during the two protocols P1 and P2 were considered optimal because, as, according to Bennett 1 , the interval between 28 and 36°C causes the reptiles to have a better functioning of metabolism.
Hellyer et al. 10 observed that the isolated administration of ketamine in greyhound dogs, produced excessive movements, violent induction and recovery, and seizures, but these effects were eliminated or minimized by joint administration of midazolam, for promoting muscle relaxation and hypnosis.Already Bienzle and Boyd 7 observed that in snapping turtles (Chelydra serpentina), the isolated use of ketamine (40 mg/kg IM) was not able to cause excitatory effects.In this study, there were no excitatory effect of ketamine on the Amazon river turtles (P.expansa) used in the experiment.
Even with the use of midazolam, there was no good muscle relaxation observed in both groups.G1 animals showed no loss of sensitivity in the thoracic and pelvic limbs and only three specimens of G2 presented it, thus showing that these protocols are not suitable for surgical procedures in P. expansa.
However, Santos et al. 11 when using midazolam as pre-anesthetical medication (2 mg/kg IM) in Phrynops geoffroanus observed good muscle relaxation, leading to easy handling and, therefore, ease of application of propofol in all Testudines.Oppenheim and Moon 12 also found good muscle relaxation inducing by applying midazolam doses of 1.5 mg/kg IM at 2.5 mg/kg IM in Trachemys scripta elegans.
The ketamine has the effect of cardiac stimulation, but its association with midazolam may reduce this effect 9 .In this study it was found that G2 showed heart rate means significantly higher (p <0.05) compared to G1 because of the higher amount of ketamine administered.Despite the use of midazolam, depending on the dose of ketamine used, it was not possible to control the heartbeat frequency of the turtles, suggesting that midazolam has limited capacity to manage these effects (Figure 1).
The specimens of G2 showed lower time for the start of immobilization, only five minutes, maintaining up to 180 minutes.
Already G1 was restrained only for twenty minutes and the animals returned movement from 60 minutes (Figure 2).This fact corroborates the findings of Bienzle and Boyd 7 , whose anesthetic association of midazolam 2 mg/kg IM with ketamine 40 mg/kg IM and midazolam 2 mg/kg IM with ketamine 20 mg/kg IM, in snapping turtles (Chelydra serpetina) showed anesthetic effects at five minutes, sedation levels remained consistent between five and twenty minutes and turtles that received higher dose showed residual sedation.According to these authors, the association with higher dose of ketamine provided a consistent and more prolonged sedation.As Carvalho and Santos 4 , the higher the dose, the faster the induction and higher the duration of anesthetic effect.
According to Holz and Holz 13 , the combination of ketamine (60 mg/kg IM) and midazolam (2 mg/kg IM), in six redeared sliders (Trachemys scripta elegans), produced an adequate anesthetic level for surgery in all Testudines studied, however there was the death of one animal.Already Bienzle and Boyd 7 stated that associations of ketamine (20 or 40 mg/kg IM) and midazolam (2 mg/kg IM), in snapping turtles (Chelydra serpentina), produced good sedation, and can thus be used only for chemical restraint of the species.In this study, while using the same association and the same doses, there was no animal death nor adequate anesthetic level for surgery observed for G2.However, Kruger and Pier 14 conducted celiotomy in Testudo hermanni hermann, and Holz and Holz 15 cardiocentesis in Trachemis scripta elegans, both just using ketamine 60 mg/kg IM.

Conclusions
Both associations were effective for sedation, but are not suitable for surgical procedures in these animals.
The Testudines submitted to protocol P2 showed faster induction and more prolonged sedative effect.
were mixed in the same syringe (3 mL) and injected intramuscularly into the left forelimb, using hypodermic needles 25 x 0.70 mm, after local antisepsis with 70% alcohol.The anesthetic parameters were measured at 0, 5, 10, 20, 30, 45, 60, 90, 120, 150 and 180 minutes after the injections.Time zero was the moment of drug administration.Subjective scores of one (1) to minimal effect, two (2) for intermediate and three (3) to optimal effect were used to evaluate the following parameters: • Locomotion: (1) animal with normal ability to move, (2) limited mobility and (3) absence of locomotion; • Muscle relaxation: (1) the animal keeps the head elevated or retracted, (2) intermediate situation and (3) the head, limbs and tail remain relaxed.For the test of pain sensitivity, the withdrawal response of the limb to clamping was considered score zero (0) and its absence, score (1).• Pain sensitivity in the right forelimb: with a curved hemostat Kelly 16 cm in the second latch on the phalanges of the hand, it is expected the response of the member's withdrawal to painful stimuli, classified as zero (0) and absence of response, one (1).• Pain sensitivity of pelvic members: procedure similar to item 4, but with clamping of the phalanges of the pelvic members;

( 3 )
for muscle relaxation and a score (1) to pain sensitivity in the right thoracic and pelvic limbs.The other seven turtles of the group achieved only score two (2) for muscle relaxation and score zero (0) to test pain sensitivity of the limbs.No studied reptile showed side effects from the drug administration.

FIGURE 1 -
FIGURE 1 -Mean heart rates at different times in Podocnemis expansa (G1 and G2) subjected to different anesthetic protocols (P1 and P2).The asterisk indicates a difference between the groups (p<0.05).

FIGURE 2 -
FIGURE 2 -Mean scores of locomotion observed at different times in Podocnemis expansa (G1 and G2) subjected to different anesthetic protocols (P1 and P2).The asterisk indicates that there were differences between groups (p<0.05).