SMALL ANIMALS
02-07-0305.qxd
1/9/2003
10:56 AM
Page 330
Postoperative hypoxemia and hypercarbia
in healthy dogs undergoing routine
ovariohysterectomy or castration and receiving
butorphanol or hydromorphone for analgesia
Vicki L. Campbell, DVM; Kenneth J. Drobatz, DVM, DACVECC, DACVIM; Sandra Z. Perkowski, VMD, PhD, DACVA
Objective—To determine frequency and severity of
postanesthetic hypoxemia and hypercarbia in healthy
dogs undergoing elective ovariohysterectomy or castration and given butorphanol or hydromorphone for
analgesia.
Design—Prospective trial.
Animals—20 healthy dogs weighing > 10 kg (22 lb).
Procedure—Dogs were anesthestized with acepromazine, glycopyrrolate, thiopental, and isoflurane, and
butorphanol (n = 10) or hydromorphone (10) was used
for perioperative analgesia. Arterial blood gas analyses were performed 10 and 30 minutes and 1, 2, 3,
and 4 hours after extubation.
Results—In dogs that received hydromorphone,
mean PaCO2 was significantly higher, compared with
the preoperative value, 10 and 30 minutes and 1, 2,
and 3 hours after extubation. Mean PaCO2 was significantly higher in dogs given hydromorphone rather
than butorphanol 10 and 30 minutes and 1 and 2
hours after extubation. Mean PaO2 was significantly
lower, compared with preoperative values, 30 minutes and 1 and 2 hours after extubation in dogs given
hydromorphone and 30 minutes after extubation in
dogs given butorphanol. Mean PaO2 was significantly
lower in dogs given hydromorphone rather than
butorphanol 1 hour after extubation. Four dogs had
PaO2 < 80 mm Hg 1 or more times after extubation.
Conclusions and Clinical Relevance—Results suggest that administration of hydromorphone to healthy
dogs undergoing elective ovariohysterectomy or castration may result in transient increases in PaCO2 postoperatively and that administration of hydromorphone
or butorphanol may result in transient decreases in
PaO2. However, increases in PaCO2 and decreases in
PaO2 were mild, and mean PaCO2 and PaO2 remained
within reference limits. (J Am Vet Med Assoc 2003;
222:330–336)
I
n 1947, Comroe and Botelho1 determined that use of
physical signs (eg, cyanosis, heart rate, respiratory
rate, and respiratory effort) was an unreliable method
of detecting mild to moderate hypoxemia in people.
The subsequent development of point-of-care pulse
oximetry made postanesthetic monitoring of O2 satu-
From the Section of Anesthesia and Critical Care, Department of
Clinical Studies, School of Veterinary Medicine, University of
Pennsylvania, Philadelphia, PA 19104.
Presented at the 2001 Annual Meeting of the American College of
Veterinary Anesthesiologists, New Orleans, October, 2001.
The authors thank Laura Gentry for assistance with anesthetic management of dogs in the study.
Address correspondence to Dr. Campbell.
330
Scientific Reports: Original Study
ration (SpO2) convenient, feasible, and reliable, and
ever since, desaturation (SpO2 < 90%) has been recognized as a common postanesthetic occurrence in people, with an incidence ranging from 14 to 16.7%.2,3 For
this reason, the American Society of Anesthesiologists
has recommended that SpO2 be monitored periodically
during the postanesthetic period.4 In addition, despite
the high cost associated with O2 supplementation, routine O2 supplementation of all postanesthetic patients
remains a frequent practice in many human hospitals.5
In veterinary medicine, postanesthetic O2 supplementation of healthy dogs is uncommon. However, little information is available in the veterinary literature
regarding the frequency and severity of postanesthetic
hypoxemia, although a pilot studya examining results
of postoperative pulse oximetry in 24 healthy dogs
found that 4 (17%) had an SpO2 < 90% during 3 evaluation periods after surgery.
Similarly, there is little information in the veterinary literature regarding the severity and frequency of
postanesthetic hypercarbia. Frequently, pure opioid
agonists are used perioperatively for analgesia in dogs,
and most pure opioid agonists cause respiratory
depression and hypercarbia.6-8 Similar to hypoxemia,
hypercarbia is difficult to detect on the basis of physical signs alone. Until a relatively high arterial partial
pressure of CO2 (PaCO2) is reached, respiratory acidosis secondary to hypoventilation may go undetected if
only physical signs are used during postanesthetic
monitoring.
The purpose of the study reported here was to
determine the frequency and severity of postanesthetic hypoxemia and hypercarbia in healthy dogs
undergoing elective ovariohysterectomy or castration. Dogs were anesthetized with acepromazine,
glycopyrrolate, thiopental, and isoflurane in O2, and
butorphanol or hydromorphone was used for perioperative analgesia. We hypothesized that these dogs
would become transiently hypoxemic and hypercarbic after surgery and, further, that hypercarbia would
be greater in dogs receiving hydromorphone, because
butorphanol is not a clinically important respiratory
depressant,9,10 whereas hydromorphone can cause
hypoventilation.11
Materials and Methods
Study population—Twenty dogs admitted to the
surgery department at the Veterinary Hospital of the
University of Pennsylvania for elective ovariohysterectomy
or castration were enrolled in the study. Dogs were included
in the study only if they weighed > 10 kg (22 lb) and results
of a physical examination, CBC, measurement of serum creJAVMA, Vol 222, No. 3, February 1, 2003
02-07-0305.qxd
1/9/2003
10:56 AM
Page 331
Anesthetic protocol—For all dogs, food was withheld
for 10 to 15 hours prior to induction of anesthesia. Water was
accessible until approximately 2 hours prior to induction of
anesthesia. All anesthetic procedures were performed by a designated certified animal health technician anesthetist. Dogs
were premedicated with acepromazine (0.02 mg/kg [0.009
mg/lb], IM) and glycopyrrolate (0.01 mg/kg [0.0045 mg/lb],
IM). The first 10 dogs received butorphanol (0.4 mg/kg [0.18
mg/lb], IM) preoperatively for analgesia, and the second 10
dogs received hydromorphone (0.2 mg/kg [0.09 mg/lb], IM).
The analgesic was administered in the same syringe as the acepromazine and glycopyrrolate, within 30 minutes prior to
anesthetic induction. The primary investigator (VLC) was not
blinded to the analgesia protocol of the dogs.
Anesthesia was induced with thiopental administered as
boluses of 2 to 4 mg/kg (0.9 to 1.8 mg/lb) to effect through
an IV catheter placed in a cephalic vein. Anesthesia was
maintained with isoflurane in O2, administered with a lowflow, closed-circuit system. All dogs received a balanced electrolyte solutionb (10 to 15 mL/kg/h [4.5 to 6.8 mL/lb/h], IV)
throughout the anesthetic period. Total anesthesia time (ie,
time from intubation until extubation), time from preoperative to postoperative opioid administration, total dose of balanced electrolyte solution administered, and total thiopental
dose were recorded.
Perioperative monitoring—For all dogs, an ECGc was
monitored continuously throughout the anesthetic period,
and indirect blood pressurec was measured periodically. An
esophageal stethoscope, temperature probe,c spirometer,d and
pulse oximetere were also used for monitoring. A side-stream
monitorf was used to measure end-tidal partial pressure of
CO2 (PETCO2), end-tidal partial pressure of isoflurane, and
inspired concentration of O2. Efforts were made to maintain
PETCO2 between 30 and 45 mm Hg, using positive pressure
ventilation (PPV) or intermittent positive pressure ventilation (IPPV) as necessary, to maintain end-tidal isoflurane concentration between 0.9 and 1.5%, and to maintain inspired O2
concentration > 90%. The sampling port for the monitor was
located between the Y-piece and the endotracheal tube
adapter. All dogs were placed on a forced-air warming deviceg
intraoperatively to maintain body temperature > 36.7oC
(98oF). Heart rate, respiratory rate, and indirect arterial blood
pressure were recorded every 5 minutes. End-tidal isoflurane
concentration, PETCO2, SpO2, and esophageal temperature
were recorded every 15 minutes. Dogs that had received
hydromorphone prior to anesthetic induction received a second dose of hydromorphone (0.1 mg/kg [0.045 mg/lb], IM)
immediately prior to extubation. Dogs that had received
butorphanol prior to anesthetic induction received a second
dose of butorphanol (0.4 mg/kg, IM) immediately prior to
extubation. Once extubated, dogs were transferred to an anesthesia recovery room, where they were placed on a circulating
warm-water blanket and allowed to breathe room air.
Arterial blood gas analyses—Preoperative samples (0.3
mL) for arterial blood gas analyses were obtained percutaneously from the dorsal metatarsal artery with a heparinized
arterial blood gas syringeh while dogs were awake, in lateral
recumbency, and breathing room air. All samples were
obtained anaerobically and analyzed at a barometric pressure
JAVMA, Vol 222, No. 3, February 1, 2003
of approximately 760 mm Hg. Immediately prior to blood gas
analysis, samples were transferred to a heparinized tube.i
Samples were analyzed with an automated devicej within 5
minutes after collection. The analyzer was maintained as recommended by the manufacturer; control samples were analyzed daily, and the analyzer self-calibrated at least hourly.
Results for all samples were temperature corrected to the
dog’s esophageal (intraoperatively) or rectal (postoperatively) temperature. Results of arterial blood gas analyses were
considered normal if the alveolar-arterial difference in partial
pressure of O2 (PAO2-PaO2) was < 15 mm Hg12 when calculated with a respiratory quotient of 0.913,14 at an inspired O2
fraction (FIO2) of 21%.
Once dogs were anesthetized, a catheter was placed in a
dorsal metatarsal artery to allow for collection of additional
arterial samples for blood gas analyses. An intraoperative
sample was obtained during stage III anesthesia, 10 to 30
minutes prior to extubation, when FIO2 was > 90%.
Additional samples were obtained 10 and 30 minutes and 1,
2, 3, and 4 hours after extubation, when FIO2 was 21%. Heart
rate, respiratory rate, rectal temperature, mucous membrane
color, pulse quality, and capillary refill time were recorded
concurrently as each sample for arterial blood gas analysis
was collected. The intraoperative and postoperative samples
for arterial blood gas analysis were obtained from the
indwelling arterial catheter while dogs were in dorsal (intraoperative) or lateral (postoperative) recumbency. At each
sampling time, 0.3 mL of arterial blood was obtained anaerobically after a 3-mL sample of blood was collected. The arterial blood sample was transferred to a heparinized tube
immediately prior to analysis. The 3-mL sample of blood was
then returned to the dog via the indwelling cephalic vein
catheter. The primary investigator was blinded to the results
of the arterial blood gas analyses until the end of the study
period for each individual dog.
Pain and sedation assessments—Pain and sedation
scores were recorded at the same time as each sample was
collected for arterial blood gas analysis. All scores were
assigned by the primary investigator, using established scoring systems.15 A numeric pain score was assigned on the basis
of degree of vocalization, movement, and agitation. The minimum score was 0, and the maximum score was 7, with a
score of 7 being the most painful. The sedation score was
assigned on the basis of arousal of the dog. The minimum
sedation score was 0, and the maximum was 2, with a score
of 2 being the most sedate. A visual analog pain score was
assigned, using a 10-cm scale with 0 being the least painful
and 10 being the most painful. At each sampling time, a mark
was made on the scale to indicate the degree of pain the dog
was exhibiting.
Statistical analyses—For continuous variables, the
Shapiro-Wilk or skewness-kurtosis test was used to determine whether data were normally distributed. Mean and SD
were used to describe continuous variables that were normally distributed; median and range were used to describe
continuous variables that were not normally distributed.
Paired Student t-tests or Wilcoxon sign rank tests (depending on data distribution) were used to compare data at various time points with preoperative data within groups.
Unpaired Student t-tests or Wilcoxon rank sum tests
(depending on data distribution) were used to compare data
between groups. For all analyses, a value of P < 0.05 was considered significant.
Results
Of the 10 dogs that received butorphanol, 4 were
female and 6 were male (1 of the males had an intraabdominal testicle); of the 10 dogs that received hydroScientific Reports: Original Study
331
SMALL ANIMALS
atinine concentration, and arterial blood gas analyses performed prior to anesthesia were normal. Signalment, body
weight, and age of all dogs were recorded. For purposes of
the present study, placement of indwelling IV and intra-arterial catheters during the study period was required. The
study protocol was approved by the Committee for the Use
of Client Owned Animals in Research at the University of
Pennsylvania. Owners of all dogs enrolled in the study signed
informed consent forms.
SMALL ANIMALS
02-07-0305.qxd
1/9/2003
10:56 AM
Page 332
morphone, 6 were female and 4 were male. Three of
the dogs that received butorphanol were Labrador
Retrievers, 2 were American Pit Bull Terriers, 1 was a
Samoyed, 1 was a Boxer, 1 was a Standard Poodle, 1
was a Rottweiler, and 1 was of mixed breeding. Six of
the dogs that received hydromorphone were of mixed
breeding, 1 was a Basset Hound, 1 was a Labrador
Retriever, 1 was a Rottweiler, and 1 was a Wirehaired
Pointing Griffon. Mean ± SD weight and age of the
dogs receiving butorphanol were 24.14 ± 8.4 kg (53.11
± 18.48 lb) and 1.9 ± 2.7 years, respectively; mean
weight and age of the dogs receiving hydromorphone
were 21.53 ± 10.26 kg (47.37 ± 22.57 lb) and 0.93 ±
0.77 years, respectively. Mean elapsed time from preoperative to postoperative opioid administration was
131.8 ± 38.0 minutes for dogs receiving butorphanol
and 187.8 ± 61.2 minutes for dogs receiving hydromorphone. Mean anesthesia time was 119 ± 36.4 minutes for dogs receiving butorphanol and 169.5 ± 64.7
minutes for dogs receiving hydromorphone. There
were no significant differences between groups in
regard to weight, age, or elapsed time from preoperative to postoperative opioid administration. Dogs
receiving hydromorphone had a significantly (P =
0.045) longer anesthesia time, compared with dogs
receiving butorphanol.
Mean preoperative arterial partial pressure of O2
(PaO2) and PaCO2 were normal for both groups of dogs
(Table 1). Mean ± SD preoperative PAO2-PaO2 was 8.03
± 4.24 mm Hg for dogs receiving hydromorphone and
5.49 ± 6.18 mm Hg for dogs receiving butorphanol.
There were no significant differences between groups
in regard to preoperative PaO2, PaCO2, or PAO2-PaO2.
Mean ± SD dose of thiopental for dogs receiving
hydromorphone was 7.46 ± 2.39 mg/kg (3.39 ± 1.09
mg/lb), and mean dose of thiopental for dogs receiving
butorphanol was 9.6 ± 3.72 mg/kg (4.36 ± 1.69 mg/lb).
Seven of the 10 dogs given hydromorphone were ventilated with a mechanical ventilator intraoperatively;
mean tidal volume was 17.45 ± 2.65 mL/kg/breath (7.9
± 1.2 mL/lb/breath). Five of the 10 dogs given butorphanol were ventilated with a mechanical ventilator
intraoperatively; mean tidal volume was 14.46 ± 0.51
mL/kg/breath (6.57 ± 0.23 mL/lb/breath). Mean endtidal isoflurane concentration throughout the anesthetic period was 1.03 ± 0.36% for dogs given butorphanol
and 0.79 ± 0.35% for dogs given hydromorphone.
Mean total volume of balanced electrolyte solution was
10.76 ± 2.26 mL/kg/h (4.89 ± 1.03 mL/lb/h) for dogs
given hydromorphone and 12.4 ± 5.94 mL/kg/h (5.64
± 2.7 mL/lb/h) for dogs given butorphanol. There were
no significant differences between groups in regard to
total thiopental dose, tidal volume, mean end-tidal
isoflurane concentration, or total volume of balanced
electrolyte solution administered.
For all dogs combined, mean ± SD intraoperative
arterial pH (FIO2 > 90%) was 7.37 ± 0.028 (range,
7.311 to 7.416), mean PaCO2 was 42.4 ± 4.1 mm Hg
(range, 35.4 to 50.2 mm Hg), mean PaO2 was 531.2 ±
83.4 mm Hg (range, 410.5 to 642.7 mm Hg), and mean
base excess was –1.09 ± 1.4 mmol/L (range, –3.5 to 1.2
mmol/L).
Among dogs that received hydromorphone, the
PaCO2 was significantly increased, compared with the
preoperative value, 10 and 30 minutes and 1, 2, and 3
hours after extubation (Table 1). In contrast, the PaCO2
was not significantly increased, compared with the preoperative value, any time after extubation in the dogs
that received butorphanol. The PaCO2 was significantly
higher among dogs that received hydromorphone than
among dogs that received butorphanol 10 and 30 minutes and 1 and 2 hours after extubation. Among 8 dogs
that received hydromorphone, 15 measurements of
PaCO2 > 45 mm Hg (range, 45.1 to 48.3 mm Hg)
occurred at all time periods except 3 hours after extubation. None of the dogs that received butorphanol
had a PaCO2 > 45 mm Hg at any time period after extubation.
Among dogs that received hydromorphone, the
PaO2 was significantly decreased, compared with the
preoperative value, 30 minutes and 1 and 2 hours after
extubation (Table 1). Among dogs that received butorphanol, the PaO2 was significantly decreased only 30
minutes after extubation. The PaO2 was significantly
lower among dogs that received hydromorphone than
among dogs that received butorphanol 1 hour after
extubation.
Among dogs that received butorphanol, the mean
PAO2-PaO2 was significantly (P = 0.028) increased,
compared with the preoperative value, 30 minutes after
extubation (9.46 ± 7.38 mm Hg). However, among
dogs that received hydromorphone, mean PAO2-PaO2
was not significantly increased any time after extubation. Mean PAO2-PaO2 was not significantly different
between groups any time after extubation.
Four dogs at 5 time periods had a PaO2 < 80 mm
Hg and a PAO2-PaO2 > 15 mm Hg. The first was a dog
given butorphanol that had a PaO2 of 78.9 mm Hg and
Table 1—Results of blood gas analyses in 20 healthy dogs undergoing routine ovariohysterectomy
and castration and given hydromorphone or butorphanol for analgesia (n = 10/group)
Variable
and group
PaCO2 (mm Hg)
Hydromorphone
Butorphanol
PaO2 (mm Hg)
Hydromorphone
Butorphanol
Postoperative values
Preoperative
value
10 minutes
30 minutes
1 hour
2 hours
3 hours
4 hours
36.9 ⫾ 2.6
38.4 ⫾ 5.0
42.3 ⫾ 4.2a
38.7 ⫾ 2.7b
43.1 ⫾ 2.6a
39.8 ⫾ 3.7b
44.6 ⫾ 3.1a
39.8 ⫾ 4.2b
42.8 ⫾ 3.1a
40.0 ⫾ 2.0b
41.0 ⫾ 2.7a
38.9 ⫾ 2.9
40.1 ⫾ 4.7
37.8 ⫾ 2.0
101.4 ⫾ 4.0
102.2 ⫾ 8.2
97.3 ⫾ 7.3
102.9 ⫾ 11.1
92.9 ⫾ 8.4a
96.7 ⫾ 7.6a
90.9 ⫾ 10.4a 95.4 ⫾ 7.7a
100.2 ⫾ 9.0b 100.3 ⫾ 6.6
96.4 ⫾ 8.5
98.3 ⫾ 8.7
97.0 ⫾ 10.1
101.3 ⫾ 4.9
Values are given as mean ⫾ SD.
a
Significantly (P ⬍ 0.05) different from preoperative value for that group. bSignificantly (P ⬍ 0.05) different from value for
dogs given hydromorphone.
332
Scientific Reports: Original Study
JAVMA, Vol 222, No. 3, February 1, 2003
02-07-0305.qxd
1/9/2003
10:56 AM
Page 333
Discussion
Results of the present study suggest that administration of hydromorphone to healthy dogs undergoing
JAVMA, Vol 222, No. 3, February 1, 2003
elective ovariohysterectomy or castration may result in
transient increases in PaCO2 postoperatively and
administration of hydromorphone or butorphanol may
result in transient decreases in PaO2. However, increases in PaCO2 and decreases in PaO2 were mild, and mean
PaCO2, mean PaO2, and mean PAO2-PaO2 remained
within reference limits, suggesting that routine postoperative O2 supplementation of healthy dogs undergoing elective ovariohysterectomy or castration is not
warranted. However, 4 dogs did have PaO2 < 80 mm Hg
1 or more times after extubation, indicating mild
hypoxemia, and 8 dogs that received hydromorphone
did have PaCO2 > 45 mm Hg 1 or more times after extubation, indicating hypoventilation.
Butorphanol, an opioid agonist-antagonist, was
chosen as 1 of the analgesics in this study, because it
has been shown to cause less respiratory depression
than pure opioid agonists9,10 and is commonly used in
veterinary practice. To the authors’ knowledge, the
respiratory depressant effects of butorphanol have not
been specifically compared to those of hydromorphone. Therefore, hydromorphone, a pure opioid
agonist, was chosen as the comparative analgesic. The
comparative potencies of these 2 analgesics are not
well understood. Butorphanol is reported to have
agonist properties 4 to 7 times as potent as those of
morphine10,16-18 and antagonist properties a fortieth of
those of naloxone.10,16 Hydromorphone is reported to
have agonist properties 4 to 10 times as potent as
those of morphine.11,18-20 Doses used in the present
study were selected on the basis of doses used clinically at the Veterinary Hospital of the University of
Pennsylvania. If equipotent doses had been used, on
the basis of butorphanol’s and hydromorphone’s relative potencies compared with morphine, then the
butorphanol dose should have been lowered to
approximately 0.2 mg/kg (0.09 mg/lb) preoperatively
and 0.1 mg/kg postoperatively. From a clinical standpoint, these are very low doses, and it did not seem
ethical to the authors to lower the dose of butorphanol for this study.
Preoperative blood gas analyses were performed in
the present study to ensure adequate oxygenation and
normocarbia prior to anesthetic induction and to allow
each dog to serve as its own control, as postoperative
values were compared with preoperative values. The
alveolar gas equation was used to determine the PAO2PaO2 for each dog while breathing room air. This gradient was used to assess oxygenation, because it
adjusts for the effects of alveolar hypoventilation.12 The
respiratory quotient is used in the equation to calculate
PAO2-PaO2 and is determined by dividing CO2 production by O2 consumption in the body.12 The value of 0.9
was chosen on the basis of reports of mean measured
respiratory quotients in healthy dogs,13 including dogs
undergoing elective surgery.14 In addition, a respiratory
quotient of 0.9 is used clinically at the Veterinary
Hospital of the University of Pennsylvania. Reports of
respiratory quotients closer to 0.7 to 0.8 have been
reported for critically ill animals,21 and the respiratory
quotient can vary as a result of many factors, including
stress, metabolic size, breed, nutritional state, hormonal rhythms, noise, and exercise state.13,14,21 Stress or
Scientific Reports: Original Study
333
SMALL ANIMALS
a PAO2-PaO2 of 25.45 mm Hg 3 hours after extubation.
The second was a dog given butorphanol that had a
PaO2 of 76.8 mm Hg and a PAO2-PaO2 of 29.2 mm Hg
10 minutes after extubation. The third was a dog given
hydromorphone that had a PaO2 of 78.3 mm Hg and a
PAO2-PaO2 of 23.41 mm Hg 30 minutes after extubation
and a PaO2 of 73.4 mm Hg and a PAO2-PaO2 of 26.77
mm Hg 1 hour after extubation. The fourth was a dog
given hydromorphone that had a PaO2 of 79.7 mm Hg
and a PAO2-PaO2 of 18.27 mm Hg 1 hour after extubation. The first, second, and fourth dogs had anesthesia
times < 100 minutes, had an elapsed time between preoperative and postoperative opioid administration
shorter than the mean time for their groups, had been
castrated, and weighed < 20 kg (44 lb). The third dog
was anesthetized for 3 hours and 25 minutes, had an
elapsed time between preoperative and postoperative
opioid administration longer than the mean time for
the hydromorphone group, had been spayed, and
weighed 41 kg (90 lb). All but the first dog had rectal
temperatures lower than the mean temperature for
their group at the time the low PaO2 was detected, and
all but the second dog were ventilated with a mechanical ventilator. Physical signs, pain scores, and sedation
scores for these dogs were not substantially different
from mean values for their groups.
Median numerical pain score ranged from 0 to 2 for
all evaluation periods in dogs given butorphanol and in
dogs given hydromorphone. Median sedation score
ranged from 0 to 1.5 for all evaluation periods for dogs
given butorphanol and from 0.5 to 2 for dogs given
hydromorphone. Median visual analog pain score
ranged from 0.7 to 1.55 for all evaluation periods for
dogs given butorphanol and from 0.9 to 1.4 for dogs
given hydromorphone. There were no significant differences between groups at any time for any of these scores.
Two hours after extubation, mean ± SD heart rate
of dogs given butorphanol (130 ± 40 beats/min) was
significantly higher than mean heart rate of dogs given
hydromorphone (99 ± 27 beats/min). Median respiratory rate ranged from 20 to 40 breaths/min for all evaluation periods in dogs given butorphanol and from 24
to 36 breaths/min in dogs given hydromorphone.
Respiratory rate was not significantly different between
groups at any time after extubation. Respiratory effort
was subjectively normal in all dogs, with occasional
panting.
Mean rectal temperature for dogs given hydromorphone was significantly lower than mean temperature
for dogs given butorphanol 10 minutes (37.1 ± 0.7oC
[98.8 ± 1.3oF] vs 37.9 ± 0.5oC [100.2 ± 0.9oF]; P =
0.015), 30 minutes (37.1 ± 0.7oC [98.8 ± 1.3oF] vs 38.1
± 0.5oC [100.5 ± 0.9oF]; P = 0.003), 1 hour (37.2 ±
0.7oC [98.9 ± 1.3oF] vs 37.9 ± 0.6oC [100.2 ± 1.0oF]; P
= 0.002), 2 hours (37.3 ± 0.7oC [99.1 ± 1.3oF] vs 38.3
± 0.4oC [101 ± 0.8oF]; P = 0.001), 3 hours (37.4 ±
0.7oC [99.3 ± 1.3oF] vs 38.3 ± 0.4oC [101 ± 0.8oF]; P =
0.002), and 4 hours (37.6 ± 0.5oC [99.6 ± 0.9oF] vs
38.3 ± 0.3oC [101 ± 0.6oF]; P = 0.002) after extubation.
SMALL ANIMALS
02-07-0305.qxd
1/9/2003
10:56 AM
Page 334
exercise can cause the respiratory quotient to approach
1.0, whereas fasting or critical illness can cause the respiratory quotient to approach 0.7. Errors imposed by
assuming a respiratory quotient higher than it actually
was would make the PAO2-PaO2 values in the present
report falsely high, and errors imposed by assuming a
respiratory quotient lower than it actually was would
make the PAO2-PaO2 values in the present report falsely low. Although the exact respiratory quotient was
unknown in this population of dogs, the calculation of
the PAO2-PaO2 remains useful in determining whether
the hypoxemia was a result of alveolar hypoventilation.
Hypoventilation is defined as alveolar ventilation
lower than that required to maintain a normal PaCO2,
given the amount of CO2 production.22-25 Clinically,
hypoventilation is determined by the PaCO2, regardless
of the PaO2. Ideally, PaCO2 values range from 35 to 45
mm Hg in dogs, although mean measured values
reported in the literature range from 33.8 to 39.8 mm
Hg.26 For the purposes of this study, a PaCO2 between
35 and 45 mm Hg was considered normal. In the present study, dogs given hydromorphone had a significantly higher PaCO2, compared with preoperative values, at all times after surgery except 4 hours after extubation. In addition, dogs given hydromorphone had a
significantly higher PaCO2 10 and 30 minutes and 1
and 2 hours after extubation than did dogs given
butorphanol. In contrast, dogs given butorphanol did
not have any significant increases in PaCO2 any time
after extubation, compared with preoperative values.
Even though PaCO2 values were significantly increased
in dogs given hydromorphone, compared with preoperative values and with values for dogs given butorphanol, mean values remained within the clinically
acceptable range at all times. Hypoventilation did
occur on an individual basis after extubation in 8 of the
dogs receiving hydromophone.
The postoperative increase in PaCO2 among dogs
given hydromorphone in the present study may have
been attributable, at least in part, to the significantly
longer total anesthesia time for this group. This could
have allowed more isoflurane to build up in the tissues,
resulting in greater respiratory depression. However,
mean end-tidal isoflurane concentrations were not significantly different between the groups, and mean endtidal isoflurane concentrations were below the published minimum alveolar concentrations for isoflurane.
Considering the respiratory depressant effects of
inhalation anesthetics are dose dependent,27 it seems
unlikely that the longer anesthetic time contributed
appreciably to the postanesthetic respiratory depression among dogs given hydromorphone.
The difference in postoperative PaCO2 values
between dogs given hydromorphone and dogs given
butorphanol could have been a result of differences in
the pharmacokinetics and pharmacodynamics of the 2
drugs. Butorphanol is completely absorbed after IM
injection and has been reported to have a duration of
effect of 1 to 2 hours for moderate pain and 2 to 4
hours for mild pain in dogs.10 The plasma half-life of
butorphanol following IM administration at a dose of
0.25 mg/kg (0.11 mg/lb) in dogs has been reported to
be 1.62 hours,28 and reported dosing intervals for
334
Scientific Reports: Original Study
butorphanol range from 1 to 4 hours.10,29
Hydromorphone is absorbed after IM injection and has
been reported to have a duration of effect of 2 to 4
hours after administration in dogs.11 Little information
about the pharmacokinetics of hydromorphone in dogs
is available in the literature, but the half-life in humans
after IV administration has been reported to be from
2.36 to 2.64 hours.30,31 Reported dosing intervals for
hydromorphone range from 2 to 6 hours.11,29 Because
specific data regarding the plasma half-life of hydromorphone in dogs have not been reported, our dosing
interval was selected on the basis of clinical usage of
the drug at the Veterinary Hospital of the University of
Pennsylvania and the commonly recommended dosing
interval of 2 to 4 hours for both drugs.10,11,29 Although
mean dosing interval was not significantly different
between groups in the present study, mean dosing
interval for dogs given butorphanol (131.8 ± 38 minutes) was shorter than the mean dosing interval for
dogs given hydromorphone (187.8 ± 61.2 minutes).
Our results, therefore, support the hypothesis that
hydromorphone is more of a respiratory depressant
than butorphanol in dogs.
For healthy dogs breathing room air at sea level,
PaO2 should typically range from 90 to 100 mm Hg,
although mean measured values in dogs reported in
the literature range from 86.5 to 97.7 mm Hg.26 For the
purposes of this study, a PaO2 between 90 and 100 mm
Hg was considered normal, and a PaO2 < 80 mm Hg,
but > 60 mm Hg, was defined as mild hypoxemia. For
dogs in both groups in the present study, mean PaO2 30
minutes after extubation was significantly lower than
the preoperative value. In addition, among dogs given
hydromorphone, mean PaO2 was also significantly
decreased 1 and 2 hours after extubation. However,
mean PaO2 was within the established reference range
at all time periods in both groups. The 2 most likely
causes of the transient decreases in PaO2 were ventilation-perfusion mismatch and alveolar hypoventilation.
The PAO2-PaO2 helps differentiate these 2 causes of
decreased PaO2. Based on the normal PAO2-PaO2 in dogs
receiving hydromorphone, the decreases in the mean
PaO2 were most likely a result of mild alveolar
hypoventilation. In dogs receiving butorphanol, the
PAO2-PaO2 was increased only 30 minutes after extubation and only in dogs given butorphanol, indicating, at
most, mild ventilation-perfusion mismatching 30 minutes after extubation in dogs given butorphanol.
In human patients, postanesthetic hypoxemia has
been attributed to abdominal surgery, obesity, prolonged duration of anesthesia, increased age, sleep
apnea, and ventilation-perfusion mismatch.32-37 In the
present study, 4 dogs at 5 different times had a PaO2 <
80 mm Hg and a PAO2-PaO2 > 15 mm Hg. This implies
that something other than alveolar hypoventilation
was causing mild hypoxemia in these 4 dogs. Whatever
the cause, it was self-correctable, in that in all dogs,
PaO2 was > 80 mm Hg within 1 to 2 hours without
external O2 supplementation. Three of these 4 dogs
weighed less than the mean weight for dogs in their
groups and had anesthesia times less than the mean
time, and none of these dogs had substantially higher
pain or sedation scores, compared with other dogs in
JAVMA, Vol 222, No. 3, February 1, 2003
02-07-0305.qxd
1/9/2003
10:56 AM
Page 335
a
Johnson JA, Murtaugh RJ. Postoperative hypoxemia by pulse oximetry in dogs (abstr), in Proceedings. 5th Annu Int Vet Emerg Crit
Care Soc Symp 1996;892.
b
Normosol-R, Abbott Laboratories, North Chicago, Ill.
c
Escort Prism Model 20403, MDE, Arleta, Calif.
d
Pediatric spirometer model 8805, Boehringer Laboratories Inc,
Norristown, Pa.
e
Biox3740 pulse oximeter, Ohmeda, Louisville, Colo.
f
POET IQ model 602, Criticare Systems Inc, Waukesha, Wis.
g
Bair hugger warming unit model 505, Augustine Medical Inc, Eden
Prairie, Minn.
JAVMA, Vol 222, No. 3, February 1, 2003
h
BD preset, reference 365423, Preanalytical Solutions, Franklin
Lakes, NJ.
i
Stat profile capillary tube kit, part No. 11533, Nova Biomedical,
Waltham, Mass.
j
Stat profile M, Nova Biomedical, Waltham, Mass.
References
1. Comroe JH, Botelho S. The unreliability of cyanosis in the
recognition of arterial anoxemia. Am J Med Sci 1947;214:1–6.
2. Morris RW, Buschman A, Warren DL, et al. The prevalence
of hypoxemia detected by pulse oximetry during recovery from anesthesia. J Clin Monit 1988;4:16–20.
3. Smith DC, Canning JJ, Crul JF, et al. Pulse oximetry in the
recovery room. Anaesthesia 1989;44:345–348.
4. American Society of Anesthesiologists Task Force on
Postanesthetic Care. Practice guidelines for postanesthetic care.
Anesthesiology 2002;96:742–752.
5. DiBenedetto RJ, Graves SA, Gravenstein N, et al. Pulse
oximetry monitoring can change routine oxygen supplementation
practices in the postanesthesia care unit. Anesth Analg 1994;78:
365–368.
6. Jacobson JD, McGrath CJ, Smith EP. Cardiorespiratory
effects of four opioid-tranquilizer combinations in dogs. Vet Surg
1994;23:299–306.
7. Pelligrino DA, Peterson RD, Henderson SK, et al.
Comparative ventilatory effects of intravenous versus fourth cerebroventricular infusions of morphine sulfate in the unanesthetized
dog. Anesthesiology 1989;71:250–259.
8. Berg RJ, Orton EC. Pulmonary function in dogs after intercostal thoracotomy: comparison of morphine, oxymorphone, and
selective intercostal nerve block. Am J Vet Res 1986;47:471–474.
9. Trim CM. Cardiopulmonary effects of butorphanol tartrate
in dogs. Am J Vet Res 1983;44:329–331.
10. Plumb DC. Butorphanol tartrate. In: Plumb DC, ed.
Veterinary drug handbook. 4th ed. Ames, Iowa: Iowa State University
Press, 2002;111–114.
11. Plumb DC. Hydromorphone. In: Plumb DC, ed. Veterinary
drug handbook. 4th ed. Ames, Iowa: Iowa State University Press,
2002;420–422.
12. West JB. Ventilation-perfusion relationships. In: West JB,
ed. Respiratory physiology—the essentials. 5th ed. Baltimore: The
Williams & Wilkins Co, 1995;51–69.
13. Walters LM, Ogilvie GK, Salman MD, et al. Repeatability of
energy expenditure measurements in clinically normal dogs by use of
indirect calorimetry. Am J Vet Res 1993;54:1881–1885.
14. Ogilvie GK, Salman MD, Kesel ML, et al. Effect of anesthesia and surgery on energy expenditure determined by indirect
calorimetry in dogs with malignant and nonmalignant conditions.
Am J Vet Res 1996;57:1321–1326.
15. Conzemius MG, Hill CM, Sammarco JL, et al. Correlation
between subjective and objective measures used to determine severity of postoperative pain in dogs. J Am Vet Med Assoc 1997;210:
1619–1622.
16. Pircio AW, Gylys JA, Cavanagh RL, et al. The pharmacology
of butorphanol, a 3,14-dihydroxymorphinan narcotic antagonist
analgesic. Arch Int Pharmacodyn Ther 1976;220:231–257.
17. Tavakoli M, Corssen G, Caruso FS. Butorphanol and morphine: a double-blind comparison of their parenteral analgesic activity. Anesth Analg 1976;55:394–401.
18. Gutstein HB, Akil H. Opioid analgesics. In: Hardman JG,
Limbird LE, Gilman AG, eds. Goodman and Gilman’s the pharmacologic basis of therapeutics. 10th ed. New York: McGraw-Hill Medical
Publishing Division, 2001;569–619.
19. Lawlor P, Turner K, Hanson J, et al. Dose ratio between morphine and hydromorphone in patients with cancer pain: a retrospective study. Pain 1997;72:79–85.
20. Hill JL, Zacny JP. Comparing the subjective, psychomotor,
and physiological effects of intravenous hydromorphone and morphine in healthy volunteers. Psychopharmacology 2000;152:31–39.
21. Walton RS, Wingfield WE, Ogilvie GK, et al. Energy expenditure in 104 postoperative and traumatically injured dogs with indirect calorimetry. J Vet Emerg Crit Care 1996;6:71–79.
22. Nunn JF. Carbon dioxide. In: Nunn’s applied respiratory
Scientific Reports: Original Study
335
SMALL ANIMALS
their groups. Two of them received butorphanol and 2
received hydromorphone. The most likely explanation
of the transient decrease in PaO2 in these dogs was ventilation-perfusion mismatch. Transient postoperative
hypoxemia may be of concern in dogs, because an
appropriate tissue O2 tension is important for proper
wound healing and prevention of wound infection.38,39
In addition, decreased postoperative tissue perfusion
may contribute to decreased tissue O2 delivery.40
Although the transient decreases in PaO2 among dogs
in the present study were apparently not detrimental, it
is possible that transient decreases in PaO2 could be
harmful to geriatric dogs and dogs with compromised
cardiorespiratory function.
The individual assigning pain and sedation scores
in the present study was not blinded to group assignment, and scores therefore may have been biased.
However, there were no significant differences between
groups in regard to numerical pain scores, sedation
scores, or visual analog pain scores. This study was not
originally intended to be a pain or sedation study.
Dogs given hydromorphone in the present study
had significantly lower rectal temperatures at all times
after anesthesia, compared with dogs given butorphanol. Dogs in this group had significantly longer
anesthesia times, compared with dogs given butorphanol, which could have contributed to hypothermia,
despite all dogs being placed on forced-air warming
devices intraoperatively and circulating warm-water
blankets postoperatively. Ambient temperatures of the
operating and anesthetic recovery rooms were not
recorded, but likely were not substantially different
between groups. It does not seem that excessive sedation could have contributed to the greater hypothermia
in the dogs given hydromorphone, because the sedation
scores were not significantly different between groups.
It is possible that hydromorphone interferes more with
thermoregulation than butorphanol. Regardless, it is
important to recognize that hypothermia may be of
concern in dogs receiving hydromorphone. Postoperative hypothermia can cause excessive shivering
and subsequently increase O2 consumption.41 However,
dogs in this study did not shiver excessively, perhaps
because the hypothermia was mild.
Results of the present study did not support routine
postanesthetic administration of O2 in healthy dogs
undergoing elective ovariohysterectomy or castration.
Future studies should be performed in geriatric dogs
and dogs with compromised cardiorespiratory function
and with additional drug protocols to determine
whether specific patient populations would benefit from
routine postanesthetic O2 supplementation or from routine postanesthetic monitoring of respiratory function.
SMALL ANIMALS
02-07-0305.qxd
1/9/2003
10:56 AM
Page 336
physiology. 4th ed. Jordan Hill, Oxford: Butterworth-Heinemann Ltd,
1993;219–246.
23. Robinson RW, Zwillich CW. Hypoventilation, central apnea,
and disordered breathing patterns. In: Bone RC, ed. Pulmonary and
critical care medicine. St Louis: Mosby Year Book Inc, 1998;Q1-1–Q18.
24. West JB. Ventilation. In: West JB, ed. Respiratory physiology—the essentials. 5th ed. Baltimore: The Williams & Wilkins Co,
1995;11–20.
25. Martin L. All you really need to know to interpret arterial
blood gases. Malvern, Pa: Lea & Febiger, 1992;17.
26. Haskins SC. Blood gases and acid-base balance: clinical
interpretation and therapeutic implications. In: Kirk RW, ed. Current
veterinary therapy VIII. Philadelphia: WB Saunders Co, 1983;
201–215.
27. Steffey EP. Inhalation anesthetics. In: Thurman JC,
Tranquilli WJ, Benson GJ, eds. Lumb & Jones veterinary anesthesia.
3rd ed. Baltimore: The Williams & Wilkins Co, 1996;297–329.
28. Pfeffer M, Smyth RD, Pittman KA, et al. Pharmacokinetics
of subcutaneous and intramuscular butorphanol in dogs. J Pharm Sci
1980;69:801–803.
29. Pascoe PJ. Opioid analgesics. Vet Clin North Am Small Anim
Pract 2000;30:757–772.
30. Parab PV, Ritschel WA, Coyle DE, et al. Pharmacokinetics of
hydromorphone after intravenous, peroral, and rectal administration
to human subjects. Biopharm Drug Dispos 1988;9:187–199.
31. Vallner JJ, Stewart JT, Kotzan JA, et al. Pharmacokinetics
and bioavailability of hydromorphone following intravenous and
336
Scientific Reports: Original Study
oral administration to human subjects. J Clin Pharmacol 1981;21:
152–156.
32. Ali J, Khan TA. The comparative effects of muscle transection and median upper abdominal incisions on postoperative pulmonary function. Surg Gynecol Obstet 1979;148:863–866.
33. Vaughan RW, Engelhardt RC, Wise L. Postoperative hypoxemia in obese patients. Ann Surg 1974;180:877–882.
34. Harte PJ, Courtney DF, O’Sullivan EG, et al. Duration of
anaesthesia and postoperative hypoxaemia. Irish J Med Sci 1982;151:
169–174.
35. Kitamura H, Sawa T, Ikezono E. Postoperative hypoxemia:
the contribution of age to the maldistribution of ventilation.
Anesthesiology 1972;36:244–252.
36. Beydon L, Hassapopoulos J, Quera MA, et al. Risk factors
for oxygen desaturation during sleep, after abdominal surgery. Br J
Anaesth 1992;69:137–142.
37. Spence AA, Alexander JI. Mechanisms of postoperative
hypoxaemia. Proc R Soc Med 1972;65:12–14.
38. Hopf H, Sessler DI. Routine postoperative oxygen supplementation. Anesth Analg 1994;79:615–616.
39. Knighton DR, Halliday B, Hunt TK. Oxygen as an antibiotic: the effect of inspired oxygen on infection. Arch Surg 1984;119:
199–204.
40. Jonsson K, Jensen JA, Goodson WH III, et al. Assessment of
perfusion in postoperative patients using tissue oxygen measurements. Br J Surg 1987;74:263–267.
41. DeWitte J, Sessler DI. Perioperative shivering: physiology
and pharmacology. Anesthesiology 2002;96:467–484.
JAVMA, Vol 222, No. 3, February 1, 2003