Journal of Zoo and Wildlife Medicine 30(1): 64-69, 1999 Copyright 1999 by American Association of Zoo Veterinarians SEVOFLURANE ANESTHESIA IN DESERT TORTOISES (GOPHERUS AGASSIZII) Matthew B. Rooney, D.V.M., Gregg Levine, D.V.M., James Gaynor, D.V.M., M.S., Ellen Macdonald, D.V.M., and Jeffrey Wimsatt, D.V.M., Ph.D. Abstract: The effects of sevoflurane on anesthesia induction, recovery, ventricular pressures, heart rate, ventricular pH, blood gas values, and electrolytes were evaluated in desert tortoises (Gopherus agassizii). Tortoises were orotra cheally intubated while awake and ventilated manually with 3-7% sevoflurane in oxygen (1 L/min) to achieve desired expired sevoflurane concentrations. Data, consisting of induction time, recovery time, systolic, diastolic, and mean ventricular pressures, heart rate, ventricular pH, blood gas values, and electrolytes, were collected prior to anesthesia and sequentially at 2.50% and 3.75% expired sevoflurane as measured at the junction of the endotracheal tube and the breathing circuit. Blood pressure was measured and blood samples were collected through a 25-ga needle passed through a cardiac access port that was placed while the tortoises were in dorsal recumbency. Mean (?SE) induction time was 2.55 ? 0.55 min, recovery time was 27.58 ? 7.55 min, and duration of anesthesia was 105 ? 12 min. Mean (?SD) values for systolic, diastolic, and mean ventricular pressures in awake tortoises were 28 ? 3 mm Hg, 22 ? 2 mm Hg, and 24 ? 2 mm Hg, respectively. Sevoflurane (2.5% expired) significantly decreased systolic (14 ? 3 mm Hg), diastolic (12 ? 1 mm Hg), and mean (13 zt 1mm Hg) ventricular pressures compared with those of awake tortoises. Ventricular pressures did not decrease further with increasing depth of anesthesia. Heart rate (32 ? 4 beats/min) did not change significantly under sevoflurane anesthesia. Sevoflurane administration increased ventricular Po2 but did not change Na+, K+, or iCa++ concentrations. Sevoflurane appears to provide safe and effective anesthesia with rapid induction and recovery. Key words: Desert tortoise, Gopherus agassizii, sevoflurane, blood gas, ventricular blood pressure, chelonians. INTRODUCTION Chelonian anesthesia is challenging. Anesthetic agents have been injected intramuscularly, intrave nously, and intracoelomically. Ketamine,430 succi nylcholine,25 midazolam,4 24 sodium pentobarbital,30 tiletamine/zolazepam,25 alfaxolone/alfadolone ace tate,25 ketamine/xylazine,13 and ketamine/midazo lam13 have been successfully used for anesthesia, sedation, and chemical restraint. However, lack of safety of some injectable anesthetics in reptiles and extremely long recovery times has limited their use 13,24,30 other potential problems with injectables include renal toxicity,12 variable individual and spe cies sensitivities,24 unpredictable absorption and metabolism, and inconsistent depth of anesthesia. Halothane and isoflurane may substantially shorten induction and recovery times19 in cheloni ans and permit better control over anesthesia, which should lead to greater survival in normal and de bilitated individuals. Chelonians can hold their breath and utilize anaerobic metabolism for extend ed periods,1 which complicates gas delivery, but en dotracheal intubation and intermittent positive pres sure ventilation help to minimize this problem.19 Sevoflurane's low blood-gas solubility coefficient permits rapid, uneventful induction, recovery, and changes in anesthesia depth.15 Sevoflurane also has little effect on tissue perfusion with oxygenated blood at clinical doses, as has been demonstrated in several species.218 For these reasons, sevoflurane may be a safe and effective anesthetic for exotic species. In this study, we determined the effects of sevoflurane on induction, recovery, ventricular pressures, heart rate, ventricular pH, blood gas val ues, and electrolytes in desert tortoises (Gopherus agassizii). MATERIALS AND METHODS Six captive desert tortoises, including four males and two females weighing 1.29-4.49 kg, were ran domly selected from a colony housed at Colorado State University, Fort Collins, Colorado. They were maintained on a 12 hr:12 hr light:dark cycle, fed grass hay and remainder grocery greens, and given water ad lib. All individuals were Mycoplasma agassizii positive, as determined by an enzyme linked immunosorbent assay, but only those ani mals that maintained body weight and were without clinical disease were included in this study. At least 1 mo prior to the study, rubber cardiac ports were implanted into the plastron, providing needle access to the heart for ventricular blood pressure determination and sampling of blood for pH, blood gas, and electrolyte analyses. To place the ports, tortoises were intubated, and anesthesia was induced and maintained with isoflurane. Tor toises were placed in dorsal recumbency and the From the Department of Clinical Sciences, College of Veterinary Medicine and Biom?dical Sciences, Colorado State University, Fort Collins, Colorado 80523, USA. 64 This content downloaded from 157.182.94.224 on Mon, 24 Mar 2014 14:14:35 PM All use subject to JSTOR Terms and Conditions
Transcript
Journal of Zoo and Wildlife Medicine 30(1): 64-69, 1999
Copyright 1999 by American Association of Zoo Veterinarians
SEVOFLURANE ANESTHESIA IN DESERT TORTOISES
(GOPHERUS AGASSIZII)
Matthew B. Rooney, D.V.M., Gregg Levine, D.V.M., James Gaynor, D.V.M., M.S., Ellen
Macdonald, D.V.M., and Jeffrey Wimsatt, D.V.M., Ph.D.
Abstract: The effects of sevoflurane on anesthesia induction, recovery, ventricular pressures, heart rate, ventricular
pH, blood gas values, and electrolytes were evaluated in desert tortoises (Gopherus agassizii). Tortoises were orotra
cheally intubated while awake and ventilated manually with 3-7% sevoflurane in oxygen (1 L/min) to achieve desired
expired sevoflurane concentrations. Data, consisting of induction time, recovery time, systolic, diastolic, and mean
ventricular pressures, heart rate, ventricular pH, blood gas values, and electrolytes, were collected prior to anesthesia
and sequentially at 2.50% and 3.75% expired sevoflurane as measured at the junction of the endotracheal tube and the
breathing circuit. Blood pressure was measured and blood samples were collected through a 25-ga needle passed through a cardiac access port that was placed while the tortoises were in dorsal recumbency. Mean (?SE) induction time was
2.55 ? 0.55 min, recovery time was 27.58 ? 7.55 min, and duration of anesthesia was 105 ? 12 min. Mean (?SD)
values for systolic, diastolic, and mean ventricular pressures in awake tortoises were 28 ? 3 mm Hg, 22 ? 2 mm Hg, and 24 ? 2 mm Hg, respectively. Sevoflurane (2.5% expired) significantly decreased systolic (14 ? 3 mm Hg), diastolic
(12 ? 1 mm Hg), and mean (13 zt 1 mm Hg) ventricular pressures compared with those of awake tortoises. Ventricular
pressures did not decrease further with increasing depth of anesthesia. Heart rate (32 ? 4 beats/min) did not change
significantly under sevoflurane anesthesia. Sevoflurane administration increased ventricular Po2 but did not change Na+,
K+, or iCa++ concentrations. Sevoflurane appears to provide safe and effective anesthesia with rapid induction and
differences in ventricular pressure mean values be
tween the two planes of sevoflurane anesthesia.
Ventricular partial pressure of oxygen (Po2), ven
tricular partial pressure of carbon dioxide (Pco2),
and blood pH were significantly influenced by sev
oflurane administered in 100% 02, but acid-base
excess (ABE) did not change (Table 2). Levels of
Po2 and pH increased whereas Pco2 levels de
creased as compared with awake values. There
were no significant differences in mean values for
pH, Po2, and Pco2 between the two planes of sev
oflurane anesthesia.
DISCUSSION
The induction time of sevoflurane anesthesia for
tortoises in this study was shorter than that of any
other inhalant anesthetic used in chelonians. Induc
tion time in this study could have been shortened
by increasing either the rate of manual ventilation
or the delivered sevoflurane concentration. An is
oflurane induction time, defined as cessation of
movement and a slowed palpebral reflex, of 7 ? 1
min (x ? SE) for Kemp's Ridley sea turtles (Lepi dochelys kempi) was reported using intermittent
positive pressure ventilation.19 Induction times of
20-50 min by mask induction in green sea turtles
(Chelonia mydas) using isoflurane also have been
reported.27 Using an open-drop method, kinosternid
turtles reached a surgical plane of anesthesia with
halothane within 11-35 min, but attempts to anes
thetize certain other species of turtles using a hal
othane open-drop method were unsuccessful.6
The mean recovery time of 27.35 ? 7.33 min in
this study was shorter than times reported in other
chelonians using other anesthetic agents. Green sea
turtles recovered from 92-452 min of isoflurane an
esthesia in 241 ? 31 min.19 Recovery was defined
as when they no longer tolerated the endotracheal
tube or were awake. It is unknown if species and
habitat of origin (desert vs. aquatic) play a role in
rate of recovery. The lower blood-gas solubility co
efficient of sevoflurane as compared with isoflurane
(0.68 vs. 1.38)15 may have conferred the properties of rapid induction, recovery, and change in depth of anesthesia, as occurs with anesthetic agents used
for humans and other mammals.15 The tortoises
were anesthetized for 105 ? 12 min. With increas
ing duration of anesthesia, recovery would proba
bly be prolonged as more anesthetic dissolved in
the tissues. The recovery protocol of ventilating
each tortoise 3 times/min on 100% 02 for 10 min allowed for rapid removal of sevoflurane. To allow
the Pco2 to increase and stimulate spontaneous
breathing, ventilation was decreased to 1 time/min.
We also used a strictly defined criterion of 30 sec of spontaneous movement for recovery to be as
sumed.
The awake heart rates in these tortoises were
slower than those reported in awake green sea tur
tles.19 Heart rates did not significantly change be
tween the awake and anesthetized states, compared with a 56% decrease in heart-rate reported in green
sea turtles under isoflurane.19 Heart rates for desert
tortoises anesthetized with other agents are un
known. Heart rate may be an insensitive measure
of anesthetic level in desert tortoises.
Although the pulse pressures measured in this
study were small, normal pressures in tortoises
have not been reported. Needle position could have
affected the measurements. If the needle had been
Table 2. Mean (?SE) pH and blood gas values in awake and anesthetized desert tortoises. ABE = acid-base
ROONEY ET AL?SEVOFLURANE ANESTHESIA IN DESERT TORTOISES 67
30H
O) I E E
W CO
o
20
? 10 c o
>
- ? SVP(mmHg)
-*? DVP(mmHg) - ?
MVP(mmHg)
Awake 2.5% Expired Sevoflurane
3.75% Expired Sevoflurane
Period of Data Collection
Figure 1. Systolic ventricular pressure (SVP), diastolic ventricular pressure (DVP), and mean ventricular pressure
(MVP) in desert tortoises. Data are expressed as x ? SE. * Significantly different from awake for all three pressures
measurements.
against a chamber wall, pulse pressure would be
artifactually decreased. Systolic, diastolic, and
mean ventricular pressures decreased from the
awake values as they did in anesthetized green sea
turtles.19 The decreased ventricular pressure in sev
oflurane-anesthetized tortoises was likely due to pe
ripheral vasodilation and to a lesser degree to
slightly decreased cardiac output, based on data
from dogs and humans.21822 Pressures did not
change from one plane of anesthesia to the next,
implying that vasodilation did not increase with
depth of anesthesia, which is consistent with find
ings in dogs anesthetized with sevoflurane.3 All
data, including that from awake animals, were col
lected with the animals in dorsal recumbency.
Therefore, positional effects on pressure readings should have been eliminated. In awake green turtles
cardiac output appears to be heart rate dependent, whereas blood pressure is heart rate independent.7 If cardiac output is predominately determined by heart rate in anesthetized desert tortoises, the find
ings that heart rate did not change with varying levels of anesthesia might indicate that cardiac out
put was minimally affected even though the blood
pressures decreased. There is no evidence that per
fusion changed significantly, based on insignificant
changes in ABE (Table 2). More information on chelonian cardiac output is needed.
The needles were placed into the heart through the plastron port blindly into either of the two atria
or the single ventricle. Based on the pressure
changes, presumably the needle was placed in the
ventricle. Ventricular, but not atrial, pressures in
mammals with four-chamber hearts decrease sig
nificantly with anesthesia.112228 Given the small
amount of variability in the pressure data from 18
cardiac punctures in six tortoises, the cardiac nee
dles were probably positioned in similar locations
in all individuals.
The values for Po2, Pco2, and pH in these tor
toises resemble those in unanesthetized green tur
tles,7 suggesting arterial sampling. Ventricular sam
pling in green-eared sliders by similar techniques resulted in blood gas values indicative of slight ve
nous mixing as compared with results obtained us
ing carotid sampling.14 The lower Po2 in desert tor
toises is most likely an effect of elevation; the study was performed at approximately 1,500 above sea
This content downloaded from 157.182.94.224 on Mon, 24 Mar 2014 14:14:35 PMAll use subject to JSTOR Terms and Conditions
venous mixture in the ventricle. This increased Pco2
would explain a slightly lower pH and may be a
reason for a slightly lower Po2.
Changes in blood gas values during anesthesia
are explained by changes in inspired oxygen con
centration and manual ventilation. Awake values
were determined while the tortoises were breathing
21% oxygen, whereas anesthetized values were de
termined with virtually 100% inspired 02. Tortoises were ventilated approximately 3 times/min during anesthesia. Variability in frequency of ventilation
and peak inspiratory pressure are most likely re
sponsible for the respiratory alkalosis that occurred
during anesthesia. The lack of a significant change
in ABE indicates there is no metabolic component
to the changes in acid-base status.
Values for Na+ and K+ determined in this study are consistent with published values. Lithium hep
arin should be used as an anticoagulant when mea
suring Na+, but sodium heparin was used in this
study. Nonetheless, any additional Na+ would have
been consistent in all samples throughout the study.
Normal iCa++ values in desert tortoises are un
known.
Expired sevoflurane was sampled from the junc
tion of the endotracheal tube and breathing circuit.
Because of the high oxygen flow rates through the
Bain circuit and the small tidal volumes, the mea
sured expired sevoflurane values were likely lower
than true expired or alveolar sevoflurane concentra
tions. Nonetheless, the relative changes are proba
bly consistent. If expired C02 had been measured,
the gradient between expired and ventricular C02
could have been used to help determine alveolar
sevoflurane concentrations. The expired sevoflurane
concentrations of 2.5% and 3.75% are slightly
higher than the values for 1.0 and 1.5 mean alveolar
concentration (MAC) of sevoflurane in mammals.26
Surgical planes of anesthesia are associated with
alveolar concentrations equivalent to approximately
1.25-1.5 MAC.21 Although the true alveolar con
centrations are probably higher than the measured
expired concentrations, it is unknown whether these
tortoises would have tolerated surgical stimulation.
With all individuals, the cloacal temperature was
assumed to be lower than the core body tempera
ture because of the anesthetic-induced relaxation of
the cloacal vent musculature, allowing ambient
room air to enter the cloaca. An esophageal tem
perature probe might be more accurate. However,
the cloacal temperature readings fall within the
lower limits of the preferred optimal temperature zone of tortoises, which is 26-38? C.1
One concern with sevoflurane use is the toxicity of Compound A (pentafluorisopropenyl fluorome
thyl ether), which is formed when sevoflurane in
teracts with C02 absorbent in closed-circuit circle
anesthesia systems. Compound A in these systems
may reach concentrations of 30-40 ppm.5 Signifi
cantly higher concentrations of Compound A are
associated with renal toxicity and death in rats.1020
To date, no clinical toxicity has been documented
in humans anesthetized with sevoflurane in low
flow circuits,9 but the manufacturer currently rec
ommends a minimum 02 flow rate of 2 L/min when
using rebreathing anesthesia circuits. The risk of
toxicity to desert tortoises and other reptiles from
Compound A is unknown but was not an issue for
this study because sevoflurane was delivered
through a nonrebreathing circuit, which uses no
C02 absorbent.
Another concern with sevoflurane, isoflurane,
and all halogenated ether anesthetics is nephrotox
icity from inorganic fluoride ions (F~). Compared with methoxyflurane, which has induced clinically significant renal disease in humans and rats, sevo
flurane and isoflurane undergo minimal intrarenal
defluorination. Nephrotoxicity is unlikely with these agents despite elevated plasma F~ levels.16
Sevoflurane-induced nephrotoxicity has not been
reported in healthy humans,23 nor has sevoflurane
exacerbated preexisting renal disease in humans.8
The degree of F~ toxicity in chelonians is unknown,
but the tortoises in this study were clinically
healthy 8 mo after anesthesia.
CONCLUSION
Sevoflurane provides rapid, controllable, and re
liable induction of anesthesia and recovery in desert
tortoises. The anesthesia period is characterized by
decreases in systolic, diastolic, and mean ventric
ular pressures but not in heart rate or Na+, K+, and
iCa++ blood concentrations.
LITERATURE CITED
1. Bennett, R. A. 1996. Anesthesia in reptiles. In:
Mader, D. (ed.). Reptile Medicine and Surgery. WB Saun
ders Co., Philadelphia, Pennsylvania. Pp. 241-247.
2. Bernard, J. M., M. F. Doursout, P. Wouters, C. J.
Hartley, R. G. Merin, and J. E. Chelly. 1992. Effects of
sevoflurane and isoflurane on hepatic circulation in the
