Veterinary Anaesthesia and Analgesia, 2005, 32, 261–270
RESEARCH PAPER
Comparative study of three intramuscular anaesthetic
combinations (medetomidine/ketamine, medetomidine/
fentanyl/midazolam and xylazine/ketamine) in rabbits
Julia Henke* DVM, Susanne Astner* DVM, Thomas Brill* DVM, MD, Barbara Eissner* DVM, Raymonde Busch  MSc
& Wolf Erhardt* Prof. DVM, Diplomate ECVA, Diplomate ECLAM
*Institute for Experimental Oncology and Therapeutic Research, Working Group Experimental Surgery, Munich, Germany
 Institute of Medical Statistics and Epidemiology, Munich, Germany
Correspondence: Julia Henke, Institut für Experimentelle Onkologie und Therapieforschung der TU München, Ismaningerstr. 22, 81675
München, Germany. E-mail: julia.henke@lrz.tu-muenchen.de
Abstract
Objective To compare the quality of surgical anaes-
thesia and cardiorespiratory effects of three intra-
muscular (IM) anaesthetic combinations in rabbits.
Study design Prospective randomized cross-over
experimental study.
Animals Nineteen adult female chinchilla mixed-
bred rabbits weighing 3.9 ± 0.8 kg.
Methods Rabbits were given one of three IM anaes-
thetic combinations: 0.25 mg kg)1 medetomidine
and 35.0 mg kg)1 ketamine (M–K), 0.20 mg kg)1
medetomidine and 0.02 mg kg)1 fentanyl and
1.0 mg kg)1 midazolam (M–F–Mz) and 4.0 mg kg)1
xylazine and 50 mg kg)1 ketamine (X–K). The effects
of anaesthesia on nociceptive reflexes, circulatory
and respiratory function were recorded. Statistical
analyses involved repeated measures ANOVA with
paired Student’s t-test applied post hoc. P-values
<0.05 were considered as significant.
Results Reflex loss was most rapid and complete in
M–K recipients, whereas animals receiving M–F–Mz
showed the longest tolerance of endotracheal
intubation (78.1 ± 36.5 minutes). Loss of righting
reflex was significantly most rapid (p < 0.05) in
the X–K group (114.7 ± 24.0 minutes). Surgical
anaesthesia was achieved in 16 of 19 animals
receiving M–K, in 14 animals receiving M–F–Mz,
and in seven animals with X–K, but only for a short
period (7.1 ± 11.6 minutes). This was significantly
(p < 0.001) shorter than with M–K (38.7 ±
30.0 minutes) and M–F–Mz (31.6 ± 26.6 minutes).
Heart rates were greatest in X–K recipients; lowest HR
were seen in animals receiving M–F–Mz. Mean
arterial blood pressure was significantly higher
(about 88 mmHg) during the first hour in the M–K
group. During recovery, the greatest hypotension
was encountered in the X–K group; minimum values
were 53 ± 12 mmHg. Six of 19 animals in the M–F–
Mz group showed a short period of apnoea (30 sec-
onds) immediately after endotracheal intubation.
Respiratory frequency was significantly lower in this
group (p < 0.001). Highest values for arterial carbon
dioxide partial pressures (PaCO2) (6.90 ± 0.87 kPa;
52.5 ± 6.5 mmHg) occurred after induction of
anaesthesia in group M–F–Mz animals. There was a
marked decrease in PaO2 in all three groups (the
minimum value 5.28 ± 0.65 kPa [39.7 ± 4.9 mmHg]
was observed with M–K immediately after injection).
Arterial PO2 was between 26.0 and 43.0 kPa (196
and 324 mmHg) in all groups during O2 delivery and
decreased – but not <7.98 kPa – on its withdrawal.
Immediately after drug injection, pHa values fell in
all groups, with lowest values after 30 minutes
(7.23 ± 0.03 with M–K, 7.28 ± 0.05 with M–F–
Mz, and 7.36 ± 0.04 with X–K). The X–K animals
261
Intramuscular anaesthetics in rabbits J Henke et al.
showed significantly (p < 0.001) higher pH values
than medetomidine recipients. During 1 hour of
anaesthesia pH values in the medetomidine groups
remained below those of the X–K group.
Conclusions Surgical anaesthesia was induced in
most animals receiving medetomidine-based combi-
nations. Arterial blood pressure was maintained at
baseline values for about 1 hour after M–K. Tran-
sient apnoea occurred with M–F–Mz and mandates
respiratory function monitoring. Oxygen enrich-
ment of inspired gases is necessary with all three
combinations. Endotracheal intubation is essential
in rabbits receiving M–F–Mz.
Clinical relevance The quality of surgical anaesthe-
sia was greatest with M–K. All combinations
allowed recoveries of similar duration. It is theoret-
ically possible to antagonize each component of the
M–F–Mz combination.
Keywords a2-agonist drugs, anaesthesia, fentanyl,
intramuscular, ketamine, medetomidine, midazo-
lam, rabbit, xylazine.
Introduction
Rabbits are sensitive to anaesthetics and so have been
generally regarded as unsafe subjects. Supporting
this, Fehr (1984) reported 5.8% deaths under xyla-
zine-ketamine anaesthesia. Rabbits are easily stres-
sed and consequently may have high plasma
catecholamine concentrations during induction of
anaesthesia, which may increase risk (Hall et al.
2001; Haberstroh & Henke 2004). Intramuscular
(IM) administration of anaesthetic combinations –
particularly xylazine/ketamine – have been popular
in the past. However, in the authors’ experience, the
analgesic properties of this combination is insufficient
for major surgery and in many cases significant
hypotension develops; this may contribute to the
reported mortality. Moreover, rabbits display wide
inter-individual variability in response to anaesthet-
ics and show strain- and sex-specific differences
(Avsaroglu et al. 2003). The most common compli-
cations during anaesthesia in small mammals are due
to respiratory depression (Sedgwick 1986; Hall et al.
2001; Haberstroh & Henke 2004). The dose of
anaesthetic required to provide surgical anaesthesia
is very close to the dose that can cause respiratory
arrest (Erhardt 1984). It is well documented that
262
most anaesthetic techniques in rabbits produce a
marked and prolonged hypotension, especially when
anaesthesia lasts longer than 1 hour (Sanford &
Colby 1980; Lipman et al. 1990; Marini et al. 1992).
Anaesthesia with combinations of agents – each
of which can be antagonized by specific antagonists
– has potential advantages. Each component may
potentiate each other’s actions, lower individual
dose requirements, and therefore produce surgical
anaesthesia more safely (Erhardt et al. 1986/87;
Henke et al. 1993, 1996, 1998, 2000; Roberts
et al. 1993). In one combination examined in the
current study, medetomidine replaced xylazine in
combination with ketamine, because its higher a2-
receptor specificity has been reported to confer
greater analgesic and sedative-hypnotic properties.
This enables the ketamine dose to be reduced
(Verstegen et al. 1991). Fentanyl is a short-acting
synthetic derivative of morphine which has potent
analgesic properties. It causes only minimal chan-
ges in circulatory variables although it can cause
marked respiratory depression. Midazolam, a ben-
zodiazepine, causes minimal haemodynamic and
respiratory changes and was chosen because of its
water solubility and therefore can be mixed with
other water-soluble substances in a single syringe.
In the present study, we investigated the anaes-
thetic and cardiorespiratory effects of three IM
anaesthetic combinations based on medetomidine
with either ketamine or fentanyl and midazolam or
xylazine with ketamine in mixed-bred chinchilla
rabbits. Another aim was to establish a completely
reversible anaesthetic technique using medeto-
midine, fentanyl and midazolam, with the aim of
i) avoiding complications associated with a long
recovery and ii) allowing rapid reversal during
emergencies. Partial antagonism of ketamine-con-
taining combinations is not recommended, partic-
ularly in the early stages of anaesthesia, because of
the undesirable effects of ketamine alone in rabbits
(Hedenqvist et al. 2002).
Materials and methods
Animals
Adult female rabbits (chinchilla, cross-bred,
n ¼ 19), with a mean body mass of 3.9 ± 0.8 kg,
were obtained from a commercial supplier (Charles
River, Sulzfeld, Germany). They were group-housed
(maximum of 10 individuals per group) with
dust-reduced wood shavings as bedding. Floor area
Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270

ranged from 0.3 to 0.5 m2 per animal. Room tem-
perature was controlled at 18–20 °C with a relative
humidity of 50–60%. A 12:12 hour light–dark
cycle was maintained, the room lighting being
automatically switched on and off at 7:00 and
19:00 hours, respectively. Animals received a
commercial pelleted diet (Altromin 2123; Hal-
tungsdiät Kaninchen, Altromin GmbH, Lage,
Germany) and water ad libitum. All animals under-
went a minimum of 10 days acclimatization period.
They were not fasted before anaesthesia.
The experiments described in this paper were
carried out with the approval of, and in accordance
with the recommendations laid down by the
‘Deutsches Tierschutzgesetz’.
Instrumentation
On the day of the experiment each animal was
weighed approximately 1 hour before physiological
variables were evaluated. Only clinically healthy
animals were accepted in the study. After weighing
and placing in a restraining cage, some hairs over
the auricular artery were removed and local
anaesthetic cream (EMLA Cream; Astra GmbH,
Wedel, Germany) was applied for 15 minutes. Fol-
lowing disinfection, a 22 SWG catheter (Venflon2;
Ohmeda, Helsingborg, Sweden) was inserted per-
cutaneously in the central auricular artery and
fixed with tape. After placing a three-way stop-cock,
an arterial line was connected to a pre-calibrated
arterial blood pressure transducer (American Opti-
cal, München, Germany) for continuous recording
of blood pressure and HR. The arterial line was also
used for repeated sampling of blood for gas analysis.
Body temperature was monitored using a rectal
probe and maintained at about 38 °C using a
heating-pad during anaesthesia.
Study protocol
After a stabilization period of about 20 minutes,
each experiment began with the recording of
pre-anaesthetic data. For the main experiment the
animals were randomly assigned to one of three
different anaesthetic techniques. Treatment order
was randomized, and each animal received all three
anaesthetics with a minimum inter-experiment
interval of 10 days. Medetomidine (Domitor; Pfizer,
Karlsruhe, Germany) or xylazine (Xylapan; Chassot,
Ravensburg, Germany) was combined with
ketamine (Narketan; Chassot) and fentanyl (Fenta-
Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270
Intramuscular anaesthetics in rabbits J Henke et al.
nyl-Janssen; Janssen, Neuss, Germany) with
midazolam (Dormicum; Hoffmann-La Roche, Grenz-
ach-Wyhlen, Germany) and medetomidine. The
three groups were: medetomidine (0.25 mg kg)1)
and ketamine (35.0 mg kg)1) (M–K); medetomidine
(0.20 mg kg)1) and fentanyl (0.02 mg kg)1) and
midazolam (1.0 mg kg)1) (M–F–Mz), or xylazine
(4.0 mg kg)1) and ketamine (50.0 mg kg)1) (X–K).
All drugs were mixed in a syringe and given as a
single IM injection into the Mm. quadriceps femoris.
As the injection volume was high (2.4–3.2 mL per
animal) doses were divided and given into both
pelvic limbs.
All animals were endotracheally intubated and
O2 was supplied for a period of 1 hour. Reflexes
were assessed and physiological variables were
recorded for an interval of 3 hours after adminis-
tering the anaesthetics. Monitoring included arterial
blood pressure, pulse oximetry, arterial blood gas
values and acid–base variables.
The following three reflexes were tested: righting
(the ability to regain sternal recumbency after
positioning on back) ear-pinch (the presence or
absence of reaction to clamping the ear margin)
pedal withdrawal (the presence or absence of pelvic
limb withdrawal on clamping the inter-digital space
between third and fourth digits). Surgical anaesthe-
sia was judged to be present if there was an absence
of response to ear-pinching and compression of the
pelvic limb nail roots.
The circulatory and respiratory variables evalu-
ated included the mean arterial blood pressure
(MAP), heart rate (HR) measured using a pulse
oximeter (Capnomac-UltimaSV; Datex/Ohmeda,
Helsingborg, Sweden) respiratory rate (fr) (based
on counting thoracic movements), blood-gas
variables (PaCO2, PaO2) and arterial pH (Synthesis
10; Instrumentation Laboratory, Regensburg,
Germany).
Time schedule
Physiological variables were recorded twice after a
10-minute stabilization period following prelimin-
ary manipulations (weighing, positioning in the
restraining cage, placing of arterial cannulae).
Time at injection of anaesthetic was defined as
t ¼ 0. Fifteen minutes after injection, endotracheal
intubation was performed using a ‘blind’ technique.
Immediately the animal was supplied with humid-
ified O2 using a face mask placed 1 cm in front of
the tube with a flow of 1 L minute)1 for exactly
263
Intramuscular anaesthetics in rabbits J Henke et al.

1 hour. The endotracheal tube was removed when
the animal began chewing and, or swallowing; this
time point was recorded.
Reflexes and physiological variables (HR, blood
pressure, respiratory frequency) were recorded
before administration of the anaesthetic (t ¼ 0),
and at 1, 3, 5, 10 and 15 minutes. After 15 min-
utes, variables were recorded at 10-minute
intervals until t ¼ 175 minutes. Arterial blood
samples were analysed at t ¼ 0, 5, 15 minutes,
and then every 15 minutes until 180 minutes after
injection.
Statistical analysis
Mean and SD for the time point of disappearance
and the duration of reflex loss were calculated. The
effects of each technique were compared initially
by using repeated measures analysis of variance
with the program SPSS (SPSS Inc., Chicago, IL,
USA, Version 11.5). Significant differences were
analysed post hoc using a paired Student’s t-test,
according to the method of Marcus et al. (1976).
Pre-anaesthetic data were compared between
treatment groups. In addition, the minimum and
maximum values of the physiological variables
were compared with the baseline (conscious) data
in each group. All the tests were performed two-
sided. p-values <0.05 were considered as sig-
nificant.
Results
All quantitative results are presented as mean ± 1
SD in Tables 1–3; time courses are shown in Figs 1–
5. Endotracheal tube tolerance (Table 1) was lon-
gest in the M–F–Mz group (78.1 ± 36.5 minutes,
p < 0.001). Extubation occurred significantly later
in animals receiving either medetomidine-based
combinations: M–K after 65.7 ± 32.7 minutes; X–K
after 24.6 ± 17.0 minutes. The righting reflex was
lost significantly faster in animals given M–K (after
1.7 ± 0.4 minutes, p < 0.01) compared with other
techniques (Table 1). The righting reflex returned
earliest in the X–K group (after 114.7 ±
24.0 minutes) which was significantly shorter
(p < 0.05) than with M–K (149.7 ± 38.7 minutes)
or M–F–Mz (129.9 ± 24.5 minutes).
The ear-pinch reflex persisted in six of 19 animals
in the X–K group, but disappeared in the remainder
for about 15 minutes. In the two other groups this
reflex was lost in nearly all animals for

Table 1 Tolerance of endotracheal intubation, time of onset, and duration of loss of reflex in rabbits anaesthetized with one
of three injectable anaesthetic techniques

M–K (1) M–F–Mz (2) X–K (3) pB pD

Tolerance of tube (n ¼ 18/19) (n ¼ 17/19) (n ¼ 16/19) p1,2 ¼ 0.215 n.s.

Duration* 65.7 ± 32.7 78.1 ± 36.5 24.6 ± 17.0 – p2,3 < 0.001
p1,3 < 0.001
Loss of righting reflex (n ¼ 19/19) (n ¼ 19/19) (n ¼ 19/19) p1,2 ¼ 0.008 p1,2 ¼ 0.04
Time of onset 1.7 ± 0.4 2.3 ± 0.8 2.0 ± 0.4 p2,3 ¼ 0.067 n.s. p2,3 ¼ 0.03
Duration* 149.7 ± 38.7 129.9 ± 24.5 p1,3 ¼ 0.320 n.s. p1,3 ¼ 0.02
Loss of ear-pinch reflex (n ¼ 18/19) (n ¼ 17/19) (n ¼ 13/19) p1,2 ¼ 0.054 n.s. p1,2 ¼ 0.151 n.s.
Time of onset 5.4 ± 4.1 12.2 ± 10.4 13.3 ± 9.4 p2,3 ¼ 0.800 n.s. p2,3 < 0.001
Duration* 52.7 ± 29.9 42.5 ± 25.8 19.6 ± 18.3 p1,3 ¼ 0.015 p1,3 < 0.001
Loss of pedal withdrawal reflex (hind) (n ¼ 19/19) (n ¼ 19/19) (n ¼ 15/19) p1,2 ¼ 0.022 p1,2 ¼ 0.027
Time of onset 4.3 ± 3.1 9.4 ± 8.0 10.9 ± 9.9 p2,3 ¼ 0.486 n.s. p2,3 < 0.001
Duration* 70.2 ± 25.5 53.5 ± 23.3 29.5 ± 23.8 p1,3 ¼ 0.018 p1,3 < 0.001
Surgical tolerance (n ¼ 16/19) (n ¼ 14/19) (n ¼ 7/19) p1,2 ¼ 0.375 n.s.
Time of onset 6.5 ± 6.0 12.1 ± 8.0 18.6 ± 9.9 n.s. p2,3 < 0.001
Duration* 38.7 ± 30.0 31.6 ± 26.6 7.1 ± 11.6 p1,3 < 0.001

Values are mean ± 1 SD in minutes. M–K, medetomidine (0.25 mg kg)1) + ketamine (35 mg kg)1) IM; M–F–Mz, medetomidine
(0.20 mg kg)1) + fentanyl (0.02 mg kg)1) + midazolam (1.0 mg kg)1) IM; X–K, xylazine (4.0 mg kg)1) + ketamine (50 mg kg)1) IM;
n ¼ number of animals whose tracheae could be intubated or showed loss of corresponding reactions/total number of animals. pB,
significant differences between groups concerning time point of loss; pD, significant differences between groups concerning duration of
reflex loss; p1,2, significant differences between M–K (1) and M–F–Mz (2); p1,3, significant differences between M–K and X–K; p2,3:
significant differences between M–F–Mz and X–K; *animals lacking reflex loss were not considered in calculating duration; n.s., not
significant.

264 Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270

 Intramuscular anaesthetics in rabbits J Henke et al.

Table 2 The effects of one of three

injectable anaesthetic techniques on M–K (1) M–F–Mz (2) X–K (3) p
heart rate and mean arterial blood
pressure in comparison with con- Heart rate (beats minute)1)
scious values (t ¼ 0) in 19 rabbits t¼0 198 ± 24 205 ± 21 203 ± 19 n.s.
Minimum 129 ± 18 115 ± 18 136 ± 20 p1,2 ¼ 0.005
p2,3 < 0.001
p1,3 ¼ 0.078 n.s.
Mean arterial pressure (mmHg)
t¼0 85 ± 8 84 ± 10 86 ± 8 n.s.
Minimum 58 ± 8 57 ± 9 53 ± 12 p1,2 ¼ 0.87 n.s.
p2,3 ¼ 0.03
p1,3 ¼ 0.04
Maximum 97 ± 8 92 ± 10 87 ± 12 p1,2 ¼ 0.06 n.s.
p2,3 ¼ 0.084 n.s.
p1,3 ¼ 0.003

Values are mean ± 1 SD. M–K, medetomidine (0.25 mg kg)1) + ketamine (35 mg kg)1)
IM; M–F–Mz, medetomidine (0.20 mg kg)1) + fentanyl (0.02 mg kg)1) + midazolam
(1.0 mg kg)1) IM; X–K, xylazine (4.0 mg kg)1) + ketamine (50 mg kg)1) IM; p, significant
differences between groups; p1,2, significant differences between M–K (1) and M–F–Mz
(2); p2,3, significant differences between M–F–Mz (2) and X–K (3); p1,3, significant
differences between M–K (1) and X–K (3); n.s., not significant.

Table 3 The effects of one of three injectable anaesthetic techniques on respiratory rate and blood-gas variables in
comparison to conscious values (t ¼ 0) in 19 rabbits

M–K (1) M–F–Mz (2) X–K (3) p

Respiratory rate (breaths minute)1)

t¼0 187 ± 24 205 ± 21 203 ± 19 n.s.
Minimum 129 ± 18 115 ± 18 136 ± 29 p1,2 < 0.001
p2,3 < 0.001
p1,3 ¼ 0.11 n.s.
PaCO2 (kPa) mmHg)
t¼0 3.85 ± 0.28 (28.9 ± 2.1) 3.95 ± 0.28 (29.6 ± 2.1) 3.91 ± 0.35 (29.3 ± 2.6) n.s.
Maximum 6.75 ± 0.73 (50.6 ± 5.5) 6.9 ± 0.87 (52.5 ± 6.5) 5.80 ± 0.69 (43.5 ± 5.2) p1,2 ¼ 0.39 n.s.
p2,3 < 0.001
p1,3 < 0.001
pH (units)
t¼0 7.48 ± 0.04 7.48 ± 0.03 7.49 ± 0.03 n.s.
Minimum 7.23 ± 0.03 7.28 ± 0.05 7.36 ± 0.04 p1,2 ¼ 0.202 n.s.
p2,3 < 0.001
p1,3 < 0.001

Values are mean ± 1 SD. M–K, medetomidine (0.25 mg kg)1) + ketamine (35 mg kg)1) intramuscular (IM); M–F–Mz, medetomidine
(0.20 mg kg)1) + fentanyl (0.02 mg kg)1) + midazolam (1.0 mg kg)1) IM; X–K, xylazine (4.0 mg kg)1) + ketamine (50 mg kg)1) IM; p,
significant differences between groups; p1,2, significant differences between M–K (1) and M–F–Mz (2); p2,3, significant differences
between M–F–Mz (2) and X–K (3); p1,3, significant differences between M–K (1) and X–K (3); n.s., not significant.

52.7 ± 29.9 minutes (M–K) and 42.5 ± 25.8 min-
utes (M–F–Mz). Differences were significant
(p < 0.001). In all animals of the M–K group, the
pedal withdrawal reflex disappeared significantly
earlier (4.3 ± 3.1 minutes) compared with the
M–F–Mz (9.4 ± 8.0 minutes) and the X–K group
(10.9 ± 9.9 minutes) p < 0.05) and remained
absent for longer periods: 70.2 ± 26.9 minutes
compared with 53.5 ± 23.3 minutes (M–F–Mz)
and 29.5 ± 23.8 minutes (X–K) p < 0.05. The
reflex was lost for approximately 30 minutes in 15
of 19 animals receiving X–K (see Table 1). The
onset of surgical anaesthesia, as indicated by the
loss of ear-pinch and nail compression reflexes, was

Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270 265

Intramuscular anaesthetics in rabbits J Henke et al.

225
Heart rate (beats minute–1)

200

175
Figure 1 Mean heart rate (beats
minute)1) in rabbits receiving one
150
of three anaesthetics. X, medetomi-
dine (0.25 mg kg)1) + ketamine
125
(35 mg kg)1) intramuscular (IM);
m, medetomidine (0.20 mg kg)1)
100
+ fentanyl (0.02 mg kg)1) + mida-
Injection t = 0
0
zolam (1.0 mg kg)1) IM; O, xylazine
–20 0 20 40 60 80 100 120 140 160 180
(4.0 mg kg)1) + ketamine (50 mg
Time (minutes) kg)1) IM.

100

90
Mean arterial blood pressure

80
(mmHg)

70
Figure 2 Mean arterial pressure
(mmHg) in rabbits receiving one of
60
three anaesthetics. X, medetomidine
(0.25 mg kg)1) + ketamine (35 mg
50 kg)1) intramuscular (IM); m, mede-
Injection t = 0 tomidine (0.20 mg kg)1) + fentanyl
0 (0.02 mg kg)1) + midazolam (1.0
–20 0 20 40 60 80 100 120 140 160 180 mg kg)1) IM; O, xylazine (4.0 mg
Time (minutes) kg)1) + ketamine (50 mg kg)1) IM.
Respiratory rate (breaths minute–1)

200

175

150
O2-supply from
125
t = 20 to t = 80
Figure 3 Mean respiratory rate in
100
rabbits receiving one of three anaes-
75 thetics. X, medetomidine (0.25 mg
50 kg)1) + ketamine (35 mg kg)1)
intramuscular (IM); m, medetomidine
25
Injection t = 0 (0.20 mg kg)1) + fentanyl (0.02 mg
0 kg)1) + midazolam (1.0 mg kg)1)
–20 0 20 40 60 80 100 120 140 160 180
IM; O, xylazine (4.0 mg kg)1) + ke-
Time (minutes) tamine (50 mg kg)1) IM.

slowest in animals of the X–K group, and its Animals receiving X–K had significantly higher
duration was significantly shorter in this group HRs (see Table 2 and Fig. 1) at all time points
(p < 0.001). Surgical tolerance was attained in compared with the other groups (p < 0.01). The
more animals in the M–K and M–F–Mz groups lowest HR was seen with the M–F–Mz combination
(16/19 and 14/19) than in the X–K group (7/19). (115 ± 18 beats minute)1). At the end of the

266 Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270

 Intramuscular anaesthetics in rabbits J Henke et al.

10.64

9.30

7.98

P a CO 2 kPa (mmHg)
6.65
Figure 4 Arterial carbon dioxide par-
tial pressure (PaCO2 kPa) in rab- 5.32
bits receiving one of three
3.99
anaesthetics. X, medetomidine
O2-supply from
(0.25 mg kg)1) + ketamine (35 mg 2.67 t = 20 to t = 80
kg)1) intramuscular (IM); m, mede-
1.33 Injection
tomidine (0.20 mg kg)1) + fentanyl t=0
(0.02 mg kg)1) + midazolam (1.0 mg 0
kg)1) IM; O, xylazine (4.0 mg –20 0 20 40 60 80 100 120 140 160 180
kg)1) + ketamine (50 mg kg)1) IM. Time (minutes)

7.60

7.50
Arterial pH

7.40
Figure 5 Arterial pH in rabbits receiv-
ing one of three anaesthetics. X,
7.30
medetomidine (0.25 mg kg)1) + ke-
tamine (35 mg kg)1) intramuscular O2-supply from t = 20 to
(IM); m, medetomidine (0.20 mg 7.20 t = 80
Injection
kg)1) + fentanyl (0.02 mg kg)1) + t=0
midazolam (1.0 mg kg)1) IM; O, 0
xylazine (4.0 mg kg)1) + ketamine –20 0 20 40 60 80 100 120 140 160 180
(50 mg kg)1) IM. Time (minutes)

observation period there were no differences be-
tween the groups. All groups showed lowest HR
from the 35th (X–K) to the 75th (M–F–Mz) minute.
The lowest HR did not differ significantly between
the M–K and X–K groups.
The medetomidine groups showed an initial
elevation in MAP (97 ± 8 mmHg in M–K and
92 ± 10 mmHg in M–F–Mz) directly after drug
injection (see Table 2 & Fig. 2). Afterwards, MAP
remained stable in the M–K group at the level of
conscious animals (approximately 85 mmHg) for
74 minutes and was higher than in other groups
(significantly higher than in X–K, p < 0.01). After
this period, MAP decreased to the level found in the
other two groups. The greatest fall in blood pressure
was seen in the X–K group, with the lowest values
(53 ± 12 mmHg) encountered during recovery
(p < 0.05).
There was a great variability in the respiratory
rate in conscious rabbits (see Table 3 & Fig. 3).
Lowest values appeared directly after endotracheal
intubation. Six of 19 animals of the M–F–Mz group
showed a short period of apnoea (30 seconds). In
this group, the respiratory frequency was signifi-
cantly (p < 0.001) lower (115 ± 18 breaths mi-
nute)1) compared with values found in the M–K
(129 ± 18 breaths minute)1) and the X–K
(136 ± 29 breaths minute)1) groups.
Values of PaCO2 in conscious animals (see
Table 3 & Fig. 4) indicated mild hypocapnia in all
groups [3.85 ± 0.28 kPa (29 ± 2 mmHg) in M–K;
3.95 ± 0.28 kPa (30 ± 2 mmHg) in M–F–Mz;
3.91 ± 0.35 kPa (29 ± 3 mmHg) in X–K]. These
values increased during anaesthesia with the high-
est values occurring in the M–F–Mz group 30 min-
utes after injection [6.9 ± 0.87 kPa
(52 ± 7 mmHg)]. This did not differ significantly
from animals receiving M–K, but was significantly
higher than the X–K group (p < 0.001). Values
were higher at most time points in the medetomi-
dine groups compared with those in the X–K group.
A substantial reduction in PaO2 was anticipated
in all three groups and so inspired breath was
routinely enriched with O2 during anaesthesia.

Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270 267

Intramuscular anaesthetics in rabbits J Henke et al.
Arterial oxygen tensions ranged from 26 to 43 kPa
(195–323 mmHg) but fell – without reaching
hypoxic values [<7.98 kPa (60 mmHg)] – after
oxygen was discontinued at the end of anaesthesia.
Pre-anaesthetic pHa values were slightly greater
than normal (see Table 3 & Fig. 5). After induction
of anaesthesia, arterial pH fell in all groups, with
lowest values being encountered after 30 minutes
(7.23 ± 0.03 in M–K recipients, 7.28 ± 0.05 in M–
F–Mz, and 7.36 ± 0.04 in animals receiving X–K).
Minimum values of pHa were more commonly
encountered in the X–K group (p < 0.001).
None of the measured conscious variables differed
significantly (p > 0.1) between groups. All minimum
and maximum values were significantly different
from data taken from conscious animals, except for
the maximum MAP values. All three anaesthetics
produced smooth and complication-free recoveries.
Most animals receiving ketamine showed chewing
movements as a sign of recovery.
Discussion
Surgery was not performed in the present study
and so the plane of anaesthesia was tested by
assessment of reflexes and the animal’s tolerance of
endotracheal intubation. The doses of X–K used
(4 mg kg)1 and 50 mg kg)1) have been described
by Harabacz (1981) as reliable and safe. In
accordance with other authors (Flecknell et al.
1983; Peeters et al. 1988; Mero et al. 1989; Hel-
lebrekers et al. 1997) the duration of surgical tol-
erance was defined as the time during which
responses to ear pinching and pelvic digit com-
pression were absent. In our study of X–K, surgical
tolerance was only achieved in a few animals, and
only for a short duration. In contrast, medetomi-
dine-ketamine combinations produced a deeper
and more prolonged reflex loss, and have enabled
cardiothoracic surgery to be performed in New
Zealand White rabbits (Difilippo et al. 2004).
Endotracheal tube tolerance was greatest in M–F–
Mz recipients because of a very prolonged loss of
laryngeal and pharyngeal reflexes and absence of
chewing reflexes. This was most probably caused
by the opioid component (fentanyl) suppressing the
cough reflex. In ketamine-based combinations, the
return of chewing movements was a good indica-
tion of recovery from anaesthesia.
The a2-agonists are known to produce marked
bradycardia; each of the three groups showed
evidence of this. The extent of HR decreases
268
encountered in the current study agrees with values
reported by others (Sanford & Colby 1980; Lipman
et al. 1990; Popilskis et al. 1991). During the first
minutes after injection, HR were lower in mede-
tomidine, compared with xylazine recipients. This
may have been due to the higher lipid solubility of
medetomidine resulting in a more rapid induction of
anaesthesia and a deeper plane of anaesthesia than
found at equivalent times in the X–K group.
Although the stimulatory cardiovascular effects of
ketamine are less marked in rabbits compared with
other species (Scabell et al. 1999) the drug was
nevertheless able to counteract the bradycardia
produced by the a2-agonists used in the current
study (Sanford & Colby 1980). Medetomidine seems
to produce a greater fall in HR at sedative doses, and
this effect may have been potentiated by fentanyl in
the M–F–Mz group, in which a maximal drop of
35% was observed compared with the rate in
conscious animals.
In our study there was a short and significant rise
in blood pressure after injection of M–K. It is notable
that MAP remained at levels found in the conscious
animals for approximately 1 hour when M–K was
used, whereas the other combinations produced
reductions below the conscious level soon after
injection. Similar results were demonstrated by
Difilippo et al. (2004) in New Zealand White rab-
bits. They reported higher blood pressures after the
use of medetomidine-, rather than xylazine-based
combinations. It should be noted that each of the
methods used in the present study decreased blood
pressure, which reached a minimum during recov-
ery, particularly in X–K animals. Therefore, partial
antagonism of ketamine-based combinations or
complete antagonism in M–F–Mz animals at the
end of surgery is recommended.
The most prominent effects on respiration were
seen in the M–F–Mz group in which fr fell markedly
and was associated with changes in all arterial
blood-gas variables. The short period of apnoea
occurring in some animals could probably be
attributed to the opioid constituent. Even lower
doses than those used in the current study have
been reported to decrease tidal volume by more than
60% in rabbits (Brown et al. 1980). The same
phenomenon has been seen in dogs when combi-
ning L-methadone with medetomidine (Scabell et al.
1999). Although the respiratory rate was severely
reduced, hypoxaemia was prevented by inspired
breath O2 supplementation. We strongly recom-
mend oxygen supplementation for all anaesthetics
Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270

in rabbits. The lowered PaCO2 encountered before
each trial in the conscious animals, and which
correlated with relatively high pH values (about
7.48) is most easily explained by stress-induced
hyperventilation. After the induction of anaesthesia
in both medetomidine groups, a mild respiratory
acidosis occurred with a maximum increase in PCO2
identified at 30 minutes after injection, when
intubation was performed and oxygen supply
was begun. The lowest pHa and highest PaCO2
values encountered here were very similar to those
measured by Hellebrekers et al. (1997) but differ
from those reported by Hedenqvist (2001) and
Hedenqvist et al. (2002) who used lower doses of
ketamine.
The longest anaesthesia time was found in the
M–K group. Lower doses of ketamine (combined
with the same medetomidine dose) – as described by
Hedenqvist (2001) proved adequate for New Zea-
land white rabbits, but the higher ketamine dose we
employed failed to produced surgical conditions in
some of the chinchilla cross-breds used in the
present study. This may reflect strain differences.
Although X–K was associated with less cardiorespi-
ratory depression than M–K, it also failed to reliably
produce surgical anaesthesia. Furthermore, it signi-
ficantly decreased blood pressure in the late anaes-
thetic period (when anaesthetic monitoring is
usually discontinued) and which could prove crit-
ical. While medetomidine appears to intensify the
analgesic and sedative effects of ketamine more than
xylazine, we recommend the following options
when IM techniques fail: i) to incorporate additional
analgesic agents like buprenorphine (Difilippo et al.
2004) or ii) to deliver low inspired concentrations of
volatile anaesthetic.
The present study points to the superiority of M
combinations over X–K anaesthesia in rabbits;
experimental studies have shown that the latter
causes severe and sustained hypotension shortly
after administration (Sanford & Colby 1980; Lipman
et al. 1990; Marini et al. 1992). The combination of
medetomidine with fentanyl and midazolam has
the advantage that surgical anaesthesia can be
produced in most rabbits without prolonging recov-
ery, because the effects of all components can be
reversed. However, this combination requires
careful monitoring of respiratory function, while
endotracheal intubation and the supplementation
of inspired gases with oxygen is strongly recom-
mended.
Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270
Intramuscular anaesthetics in rabbits J Henke et al.
Acknowledgements
The authors wish to thank Professor Paul Flecknell
for his help in reviewing this manuscript.
The study was partly supported by the ‘Arbeits-
kreis für Transfusionsmedizin, Onkologie und
Hämostaseologie’ and by Pfizer company (Karls-
ruhe, Germany) in providing the medetomidine.
References
Avsaroglu H, Versluis A, Hellebrekers LJ et al. (2003)
Strain differences in response to propofol, ketamine and
medetomidine in rabbits. Vet Rec 152, 300.
Brown JH, Pleuvry BJ, Kay B (1980) Respiratory effects of
a new opiate analgesic, R 39209, in the rabbit: com-
parison with fentanyl. Br J Anaesth 52, 1101–1106.
Difilippo SM, Norberg PJ, Suson UD et al. (2004) A com-
parison of xylazine and medetomidine in an anaesthetic
combination in New Zealand White rabbits. Contemp
Top Lab Anim Sci 43, 32–34.
Dupras J, Vachon P, Cuvelliez S et al. (2001) Anesthesia of
the New Zealand rabbit using the combination of tile-
tamine-zolazepam and ketamine-midazolam with and
without xylazine. Can Vet J 42, 455–460.
Erhardt W (1984) Anästhesieverfahren beim Kaninchen.
Tierärztl Prax 12, 391–402.
Erhardt W, Stephan M, Schatzmann U et al. (1986/87)
Immediate and complete reversal of anaesthesia by sim-
ultaneous administered benzodiazepine antagonists and
opioid antagonists in the dog. J Vet Anaesth 14, 90–99.
Fehr M (1984) Narkose bei Heimtieren. Kleintierprax 29,
313–318.
Flecknell PA, John M, Mitchell M et al. (1983) Neuro-
leptanalgesia in the rabbit. Lab Anim 17, 104–109.
Haberstroh J, Henke J (2004) Kaninchen. In: Anästhesie
und Analgesie beim Klein- und Heimtier (1st edn). Er-
hardt W, Henke J, Haberstroh J (eds). Schattauer,
Stuttgart, Germany, pp. 629–641.
Hall LW, Clarke KW, Trim CM (2001) Veterinary Anaes-
thesia (10th edn). WB Saunders, London, UK.
Harabacz I (1981) Die Neuroleptanalgesie mit Ketamin-
Xylazin beim Kaninchen. Vet Med thesis, LMU Munich.
Hedenqvist P (2001) Assessment of ketamine/medetomi-
dine anaesthesia in the New Zealand White rabbit. Vet
Anaesth Analg 28, 18–25.
Hedenqvist P, Orr HE, Roughan HE et al. (2002) Anaes-
thesia with ketamine/medetomidine in the rabbit:
influence of route of administration and the effect of
combination with Butorphanol. Vet Anaesth Analg 29,
14–19.
Hellebrekers LJ, de Boer EJW, van Zuylen MA et al. (1997)
A comparison between medetomidine-ketamine and
medetomidine-propofol anaesthesia in rabbits. Lab Anim
31, 58–69.
269
Intramuscular anaesthetics in rabbits J Henke et al.
Henke J, Roberts U, Otto K et al. (1993) Fully antagoniz-
able anaesthesia of the guinea pig. Part II: Clinical
investigations. In: Neue Aspekte der Veterinäranästhesie
und Intensivtherapie (1st edn). Schmidt-Oechtering G,
Alef M (eds). Paul Parey, Berlin, Germany, p. 296.
Henke J, Brill T, Haberstroh J et al. (1995) General
anaesthesia in several mammal species. Tier-
schutzbeauft 2, 171–187.
Henke J, Roberts U, Otto K et al. (1996) Klinische Unter-
suchungen zur i.m. Kombinationsanästhesie mit Fenta-
nyl/Climazolam/Xylazin und postoperativer i.v.
Antagonisierung mit Naloxon/Sarmazenil/Yohimbin
beim Meerschweinchen. Tierärztl Prax 24, 85–87.
Henke J, Lendl C, Mantel R et al. (1998) Reversal of
anaesthesia in rats: effects on various variables. Pro-
ceedings of the Spring meeting AVA, Edinburgh, p. 70.
Henke J, Schneider E, Erhardt W (2000) Medetomidine
combination anaesthesia with and without antagoni-
sation- influence on vital variables in mongolian gerbils
(mesocricetus unguiculatus). Proceedings of the 7th
WCVA, Berne, Switzerland, pp. 99–100.
Lipman NS, Marini RP, Erdman SE (1990) A comparison
of ketamine/xylazine and ketamine/xylazine/aceproma-
zine anesthesia in the rabbit. Lab Anim Sci 40, 395–
398.
Marcus R, Peritz E, Gabriel KR (1976) On closed testing
procedures with special reference to ordered analysis of
variance. Biometrika 63, 655–660.
Marini RP, Avison DL, Corning BF et al. (1992) Ketamine/
xylazine/butorphanol: a new anesthetic combination for
rabbits. Lab Anim Sci 42, 57–62.
270 Ó Association of Veterinary Anaesthetists, 2005, 32, 261–270
Mero MS, Vainionpää S, Vasenius J et al. (1989) Mede-
tomidine-ketamine-diazepam anaesthesia in the rabbit.
Acta Vet Scand 85(Suppl.), 135–137.
Peeters ME, Gil D, Teske E et al. (1988) Four methods for
general anaesthesia in the rabbit: a comparative study.
Lab Anim 22, 355–360.
Popilskis SJ, Oz MC, Gorman P et al. (1991) Comparison of
xylazine with tiletamine-zolazepam (Telazol) and xyla-
zine-ketamine anesthesia in rabbits. Lab Anim Sci 41,
51–53.
Roberts U, Henke J, Brill T et al. (1993) Fully antagoniz-
able anaesthesia of the guinea pig. Part I: Experimental
investigations. In: Neue Aspekte der Veterinäranästhesie
und Intensivtherapie (1st edn). Schmidt-Oechtering G,
Alef M (eds). Paul Parey, Berlin, p. 295.
Sanford TD, Colby ED (1980) Effect of xylazine and keta-
mine on blood pressure, heart rate and respiratory rate
in rabbits. Lab Anim Sci 30, 519–523.
Scabell P, Henke J, Deppe H et al. (1999) Vergleichsun-
tersuchungen zur Medetomidin-Kombinationsanästhe-
sie beim Hund. Tierärztl Prax 27, 231–238.
Sedgwick CJ (1986) Anesthesia for rabbits. Vet Clin North
Am Food Anim Pract 2, 731–736.
Verstegen J, Fargetton X, Donnay I et al. (1991) An
evaluation of medetomidine/ketamine and other drug
combinations for anaesthesia in cats. Vet Rec 128, 32–
35.
Received 5 September 2003; accepted 3 October 2004.