Veterinary Surgery

30:351-358, 2001

Analgesic Effect of the Transdermal Fentanyl Patch During and

After Feline Ovariohysterectomy

LEIGH E. GLERUM, DVM, CHRISTINE M. EGGER, DVM, MVSc, Diplomate ACVA,

SHEILA W. ALLEN, DVM, MS, Diplomate ACVS, and MICHELLE HAAG, BS

Objective—To evaluate the efficacy of the transdermal fentanyl patch in relieving perioperative pain
and stress associated with ovariohysterectomy in cats.
Study Design—Prospective laboratory trial.
Animals—Twenty-four female, purpose-bred cats.
Methods——Each cat was randomly assigned to groups 1-3. Group 1 received a 25-␮g/h
transdermal fentanyl patch only. Group 2 received the patch and anesthesia. Group 3 received
anesthesia only. Patches were left in place for 72 hours. Rectal temperature, heart rate, respiratory
rate, indirect blood pressure, blood glucose, serum cortisol concentration, plasma fentanyl concen-
tration, pain score, and excitement/sedation score were monitored at prescribed intervals over an
81-hour period. Cats from groups 1-3 were reassigned to groups 4 and 5. Group 4 received the patch,
anesthesia, and an ovariohysterectomy. Group 5 received anesthesia and an ovariohysterectomy only.
The study period and monitored parameters were the same as for groups 1-3.
Results—Serum cortisol concentrations were significantly lower in group 4 than group 5 during the
surgical and early postsurgical time periods. A similar effect was noted in blood glucose
concentrations during the surgical period. Rectal temperature was significantly higher in group 2
when comparing all anesthetized groups during the early postsurgical period. Pain scores were
significantly higher in groups 4 and 5 than in groups 2 and 3 during the early postsurgical period.
There was no significant difference in pain scores between groups 4 and 5 during this period,
however.
Conclusions—The transdermal fentanyl patch affects biochemical markers of perioperative pain and
stress associated with ovariohysterectomy in cats, attenuating rises in serum cortisol and blood
glucose concentrations during the surgical and early postsurgical periods.
Clinical Relevance—The transdermal fentanyl patch is effective in alleviating perioperative pain and
stress associated with ovariohysterectomy in cats as evidenced by attenuated rises in cortisol and
blood glucose concentrations in cats that were operated on and treated with the patch.
© Copyright 2001 by The American College of Veterinary Surgeons

P AIN MANAGEMENT is a rapidly developing

discipline in veterinary medicine and is of partic-
ular importance to veterinary surgeons. The physio-
logic responses to perioperative pain and stress can be
deleterious to the healing process.1-3 A myriad of
analgesic agents are available to the clinician. How-
ever, effective use of these drugs is contingent on pain
recognition and the selection of the appropriate treat-
ment.
Perioperative pain management in cats is particu-
larly challenging because of the unique behavioral
responses of the species. Cats tend to be stoic in

From the Department of Small Animal Medicine, College of Veterinary Medicine, University of Georgia, Athens, GA.
Supported by an ACVS Surgeon-in-Training grant and The University of Georgia Veterinary Medical Experiment Station.
No reprints available.
Address correspondence to Leigh E. Glerum, DVM, Diplomate ACVS, Veterinary Surgical Associates, 251 North Amphlett Blvd, San
Mateo, CA 94401.
© Copyright 2001 by The American College of Veterinary Surgeons
0161-3499/01/3004-0007$35.00/0
doi:10.1053/jvet.2001.24387

351
352 TRANSDERMAL FENTANYL PATCH
painful postoperative situations, often exhibiting pain
via subtle signs such as postural changes,1,4 as op-
posed to more readily recognized manifestations such
as vocalization.5 Choosing an analgesic drug for the
feline patient also poses a particular challenge. Be-
cause nonsteroidal anti-inflammatory drugs must be
used with extreme caution in cats, opiates are gener-
ally preferred. However, dose-dependent adverse re-
actions such as excitement or dysphoria may be
exhibited by cats receiving opiates.6,7
The transdermal fentanyl patch enables the clinician
to provide opioid analgesia for several days of the
perioperative period, requiring only one nonpainful
administration. Multiple studies have characterized the
pharmacokinetics8-10 and evaluated the efficacy11-13
of transdermal fentanyl in dogs. One published report
addressed the use of transdermal fentanyl in cats,
primarily describing pharmacokinetics in nonoperated,
awake cats. Significant respiratory depression or
changes in behavior, appetite, and fecal or urinary
output did not occur while the patches were in place.14
Another report described attainment and long-term
maintenance of blood concentrations of fentanyl in
nonoperated, awake cats after transdermal administra-
tion that was presumed to be adequate for analgesia
based on reports of fentanyl concentrations necessary
for analgesia in dogs and humans.15 The minimal
alveolar concentration of isoflurane was significantly
reduced in nonoperated, anesthetized cats by the pres-
ence of both 25- and 50-␮g/h fentanyl patches. Ad-
verse side effects were noted with the higher-concen-
tration patch, including anorexia and constipation.16
The purpose of this study was to determine the
efficacy of the transdermal fentanyl patch for relieving
perioperative pain and stress in cats undergoing ovar-
iohysterectomy.
MATERIALS AND METHODS
Twenty-four female purpose-bred cats were obtained
through the University of Georgia laboratory animal re-
sources. The experimental protocol was approved by the
University of Georgia Animal Care and Use Committee,
and husbandry was provided according to established guide-
lines. The mean age was 11.7 months (range, 11-14
months). The mean body weight was 3.0 kg (range, 2.3-4.3
kg). All cats were determined to be normal by physical
examinations and measurement of packed cell volume and
plasma protein, blood urea nitrogen, and blood glucose
concentrations. Cats were assigned randomly to group 1, 2,
or 3.
Group 1: Transdermal Fentanyl Patch Only (8 cats)
Venous catheters (18-gauge, 8-inch; I-Cath; Chartermed,
Inc, Lakewood, NJ) were placed in either a jugular or
medial saphenous vein a minimum of 12 hours before
transdermal fentanyl patch placement. To aid with catheter
placement, cats were sedated with a single intramuscular
injection of 10 mg/kg ketamine HCl (Ketaset; Fort Dodge
Animal Health, Fort Dodge, IA), 0.05 mg/kg acepromazine
maleate (Fermenta Animal Health Co, Kansas City, MO),
and 0.04 mg/kg atropine sulfate (Phoenix Scientific, Saint
Joseph, MO). Future patch placement sites on either the
right or left lateral hemithorax were shaved.
A 25-␮g/h transdermal fentanyl patch (Duragesic; ALZA
Corporation, Palo Alto, CA) was placed on the lateral
thorax and covered with a light wrap (encompassing the
patch and thorax) at time 0. Simple descriptive scores17,18
for pain and excitement/sedation (Table 1) were assigned by
1 of 2 observers, and respiratory rate was recorded before
cat handling. Heart rate, indirect blood pressure (via ultra-
sonic doppler probe [Ultrasonic Doppler Flow Detector;
Parks Medical Electronics Inc, Aloha, OR] and pneumatic
cuff [Critikon Cuff; Johnson & Johnson Medical Inc,
Arlington, TX] attached to a mercury sphygmomanometer
[Tycos; Welch Allyn Company, Arden, NC]), and rectal
temperature were recorded. Blood samples were obtained
from the venous catheter for measurement of concentrations
of serum cortisol (1 mL whole blood placed in a clot tube),
plasma fentanyl (1 mL whole blood placed in a tube
containing ethylenediaminetetraacetic acid), and blood glu-
cose (1 drop whole blood placed immediately onto a
glucometer [Accu-chek Advantage; Boehringer Mannheim
Corporation, Indianapolis, IN] testing strip). Six milliliters
of 0.9% NaCl (McGaw, Inc, Irvine, CA) was administered
Table 1. Simple Descriptive Scoring System
Simple Descriptive Score for Excitement/Sedation
⫺1 Unarousable
Moderately sedated, but rousable
Mildly sedated
0 No sedation or excitement
⫹1 Restless; calms with petting or holding
Agitated, vocalizing, moving about in cage; no response to
petting or holding
Thrashing, rolling in cage, clawing at bedding
Simple Descriptive Score for Pain
1 No pain, allows palpation of abdomen and/or surgical site,
normal posture
2 Faint pain, attention to but allows palpation of abdomen and/
or surgical site, tucked up when standing
3 Mild pain, withdraws from or vocalizes in response to
palpation, hunched body position in sternal recumbency,
paws tucked beneath body
4 Moderate pain, tries to escape in response to palpation,
frequent body-position changes, restless
5 Extreme pain, intensely still to writhing
 GLERUM ET AL
intravenously to compensate for the blood volume removed
(3 times volume extracted). The catheter was then flushed
with 1 mL of heparinized saline. Subsequent recordings
of pain score, excitement/sedation score, respiratory rate,
heart rate, indirect blood pressure, and rectal temperature
were made at times 2, 4, 6, 8, 12, 13.5, 14, 15, 16, 17, 25,
37, 49, 61, and 73 hours. Subsequent blood samples were
obtained for measurement of concentrations of serum
cortisol at times 4, 8, 12, 14, 15, 16, 17, 25, 37, 49, 71, and
73 hours; plasma fentanyl at times 2, 4, 6, 8, 12, 13.5, 15,
17, 25, 37, 49, 61, and 73 hours, and blood glucose at times
2, 4, 6, 8, 12, 13.5, 14, 15, 16, 17, 25, 37, 49, 61, and 73
hours.
The fentanyl patch was removed after sampling at time
73 hours. Recordings of pain score, excitement/sedation
score, respiratory rate, heart rate, indirect blood pressure,
and rectal temperature were made at times 74, 75, 77, and
81 hours. Blood samples were obtained for measurement of
concentrations of serum cortisol at times 77 and 81 hours;
plasma fentanyl at times 74, 75, 77, and 81 hours; and blood
glucose at times 74, 75, 77, and 81 hours.
All blood samples were stored under refrigeration until
being centrifuged for serum or plasma extraction. Samples
were then stored at ⫺70°C until assays were performed.
Radioimmunoassays were used to measure both serum
cortisol (Coat-A-Count Cortisol kit; Diagnostic Products
Corporation, Los Angeles, CA) and plasma fentanyl con-
centrations (Fentanyl RIA kit; Janssen Biotech NV, Olen,
Belgium).
Group 2: Transdermal Fentanyl Patch and
Anesthesia (8 cats)
Cats in group 2 were treated identically to group 1 before
anesthesia.
The anesthesia sequence began at time 11.5 hours.
Premedication of 5 mg/kg ketamine, 0.05 mg/kg aceproma-
zine, and 0.04 mg/kg atropine were all administered intra-
muscularly. A catheter (22-gauge, 2-inch; Sovereign
Indwelling Catheter; Sherwood Medical Industries, Tul-
lamore, Ireland) was placed in a cephalic vein for adminis-
tration of anesthetic induction drugs and intravenous fluid
support. Anesthesia was induced with 8 mg/kg ketamine 30
minutes after premedication. Anesthesia was maintained for
30 minutes with halothane and oxygen administered via an
endotracheal tube and rebreathing circuit. Lactated Ringer’s
solution (McGaw, Inc) was administered intravenously
throughout the anesthetic period at 10 mL/kg/h. The fol-
lowing time points correspond with perianesthetic events:
12 hours ⫽ just before induction; 12.5 hours ⫽ postinduc-
tion; 13 hours ⫽ termination of halothane administration;
13.5 hours ⫽ extubation; and 14 hours ⫽ 1 hour following
cessation of halothane administration.
353
Recordings of pain score, excitement/sedation score,
respiratory rate, heart rate, indirect blood pressure, rectal
temperature, and blood sampling for serum cortisol, plasma
fentanyl, and blood glucose concentration measurement
were identical to group 1, with the following additions: pain
score, excitement/sedation score, respiratory rate, heart rate,
indirect blood pressure, and rectal temperature recorded at
time 12.5 hours; serum cortisol samples obtained at times
12.5 and 13.5 hours; blood glucose sample obtained at time
12.5 hours.
Group 3: Anesthesia Only (8 cats)
Venous catheters for blood sampling were placed as
described for group 1. A light wrap was placed around the
thorax at time 0. Simple descriptive scores for pain and
excitement/sedation were assigned, and respiratory rate was
recorded before cat handling. Heart rate, indirect blood
pressure, and rectal temperature were recorded. Blood
samples were obtained from the catheter for measurement
of serum cortisol (1 mL whole blood placed in a clot tube)
and blood glucose (1 drop whole blood placed immediately
onto a glucometer testing strip) concentrations. Three mil-
liliters of 0.9% NaCl was administered intravenously to
compensate for the blood volume removed (3 times volume
extracted). The catheter was then flushed with 1 mL of
heparinized saline.
The anesthesia sequence, subsequent monitoring, and
subsequent blood sampling were as described for group 2,
with the following exceptions: plasma fentanyl concentra-
tion was not measured at any time; blood glucose concen-
tration was not measured at times 2 or 6 hours.
The thoracic wrap was removed after sampling at time 73
hours. Recordings of pain score, excitement/sedation score,
respiratory rate, heart rate, indirect blood pressure, and
rectal temperature were made at times 74, 75, 77, and 81
hours. Blood samples were obtained for measurement of
serum cortisol and blood glucose concentrations at times 77
and 81 hours.
Cats in groups 1, 2, and 3 were reassigned to groups 4 and
5 such that there was equal distribution of group 1, 2, and 3
cats in the resulting 2 groups. A minimum of 1 week was
allowed between trials.
Group 4: Transdermal Fentanyl Patch, Anesthesia,
and Ovariohysterectomy (12 cats)
Cats in group 4 were managed preoperatively as de-
scribed for group 1, and a 25-␮g/h transdermal fentanyl
patch was placed on the lateral thorax and covered with a
light wrap at time 0.
The anesthesia/surgery sequence began at time 11.5
hours, at which time the cats were anesthetized as described
for group 2. Anesthesia was maintained for 1 hour with
354 TRANSDERMAL FENTANYL PATCH
halothane and oxygen administered via an endotracheal tube
and rebreathing circuit. Lactated Ringer’s solution was
administered intravenously throughout the anesthetic period
at 10 mL/kg/h. Ovariohysterectomy was performed during
the general anesthetic period by the same surgeon in all cats.
The following time points correspond with perianesthetic
events: 12 hours ⫽ just before induction; 12.5 hours ⫽
postinduction; 13 hours ⫽ termination of halothane admin-
istration; 13.5 hours ⫽ extubation; and 14 hours ⫽ 1 hour
following cessation of halothane administration.
Recordings of pain score, excitement/sedation score,
respiratory rate, heart rate, indirect blood pressure, rectal
temperature, and blood sampling for measurement of serum
cortisol, plasma fentanyl, and blood glucose concentrations
were identical to group 2, with the following additions: pain
score, excitement/sedation score, respiratory rate, heart rate,
indirect blood pressure, and rectal temperature recorded at
time 13 hours; and blood samples obtained for measurement
of serum cortisol and blood glucose concentrations at time
13 hours.
The fentanyl patch was removed after sampling at time
73 hours, and subsequent monitoring and sampling were as
described in group 2.
Group 5: Anesthesia and Ovariohysterectomy
(12 cats)
Cats in group 5 were managed preoperatively as de-
scribed for group 3.
The anesthesia/surgery sequence began at time 11.5
hours as described for group 2. Anesthesia was maintained
for 1 hour with halothane and oxygen administered via an
endotracheal tube and rebreathing circuit. Lactated Ringer’s
solution was administered intravenously throughout the
anesthetic period at 10 mL/kg/h. Ovariohysterectomy was
performed during the general anesthetic period by the same
surgeon as in group 4. The time points corresponding with
perianesthetic events and the subsequent monitoring and
sampling were identical to group 4, with the following
exceptions: plasma fentanyl concentrations were not mea-
sured; blood glucose concentration was not measured at
times 2 and 6 hours.
The thoracic wrap was removed after sampling at time 73
hours, and subsequent monitoring and sampling were as
described in group 3.
Mean surgical time was 27 minutes (range, 20-40 min-
utes). Patches did not loosen prematurely in any cats, nor
did any have to be removed early because of adverse side
effects. No supplemental analgesia was administered to
either surgical group. Cats were reassigned to a long-term
nutritional study of spayed female cats following comple-
tion of this investigation.
A statistical mixed model was used to determine if
fentanyl concentrations significantly affected cortisol con-
centrations among the individuals of group 1 over time
periods 0-81 hours and 0-37 hours. A repeated-measures
ANOVA was used to determine whether there were signif-
icant effects due to group (groups 1-3; groups 4 and 5) on
measurements during the time periods 0-37 hours and 14-25
hours. A repeated-measures ANOVA was used to determine
whether there were significant effects due to group (groups
4 and 5) on blood pressure, heart rate, blood glucose, and
cortisol when comparing specific time periods: presurgery
(PS) 0-8 hours, surgery (S) 12-13.5 hours, early postsurgery
(EPS) 14-25 hours, and late postsurgery (LPS) 37-73 hours.
A one-way ANOVA was used to determine whether there
were any differences among the groups (groups 2-5) with
regard to treatments and treatment interactions on measured
physiologic parameters during the EPS period. Mean pain
scores and mean excitement/sedation scores were compared
among groups (groups 2-5) during the EPS period using a
1-way ANOVA. Significance was set at P ⬍ .05. Further
analysis of the fentanyl pharmacokinetic data will be pre-
sented in a companion article.
RESULTS
Mean cortisol concentrations were significantly
higher in groups 4 and 5 during the surgical and early
postsurgical time periods when compared with all
other time periods combined (P ⫽ .0001 and .0003,
respectively). The group 4 mean cortisol concentration
(3.3 ␮g/dL) was significantly less than the group 5
concentration (4.7 ␮g/dL) during the surgical time
period (P ⫽ .0002). The group 4 mean cortisol
concentration (3.7 ␮g/dL) also was significantly less
than the group 5 concentration (5.5 ␮g/dL) during the
early postsurgical period (P ⫽ .0448) (Fig 1). When
Fig 1. Mean serum cortisol concentrations for surgical
groups (4 and 5) during specific time periods presurgery (PS,
0-8 hours), surgery (S, 12-13.5 hours), early postsurgery (EPS,
14-25 hours), and late postsurgery (LPS, 37-73 hours). *Mean
cortisol concentrations were statistically different between
groups.
 GLERUM ET AL 355

all anesthetized groups were analyzed for the effect of

treatment on cortisol concentration during the early
postsurgical period, the application of surgery (P ⫽
.0221) and the application of surgery with the fentanyl
patch in place (P ⫽ .0384) were both found to be
significant factors. A higher mean cortisol concentra-
tion was found in the group receiving surgery without
the patch in place (group 5 mean ⫽ 5.5 ␮g/dL) than in
any other anesthetized group (mean ⫽ 3.7 ␮g/dL, 2.7
␮g/dL, and 3.7 ␮g/dL for groups 2, 3, and 4, respec-
tively) (Fig 2).
Mean blood glucose concentrations were signifi-
cantly higher in groups 4 and 5 during the surgical and
Fig 3. Mean blood glucose concentrations for anesthetized
early postsurgical time periods when compared with groups (2-5) for time 8-37 hours, including surgical (12-13.5
all other time periods combined (P ⫽ .0173 and .0001, hours) and early postsurgical (14-25 hours) periods.
respectively). The mean blood glucose concentration
for group 4 (75 mg/dL) was significantly less than for
group 5 (93 mg/dL) during the surgical period (P ⫽ found in group 2 (102.8°F) than in all other anesthe-
.0031). The occurrence of surgery had a significant tized cats (mean temperature ⫽ 101.2°F, 101.1°F, and
effect on blood glucose concentrations during the early 101.0°F in groups 3, 4, and 5, respectively) (Fig 4).
postsurgical period (P ⫽ .0009). Both surgical groups Rectal temperature was the only measurement affected
(4 and 5) had higher mean blood glucose concentra- by group when evaluating groups 1 through 3 sepa-
tions than the nonsurgical groups (mean blood glu- rately from groups 4 and 5. When evaluating groups 1
cose ⫽ 88 mg/dL, 86 mg/dL, 106 mg/dL, 116 mg/dL through 3, temperature was found to be significantly
in groups 2 through 5, respectively) (Fig 3). The blood different over time period 0-37 hours (P ⫽ .0003).
glucose concentrations remained within normal limits Specifically, rectal temperature was higher in group 2
in all groups during the late postsurgery period (after (mean ⫽ 102.4°F) than group 3 (mean ⫽ 101.3°F).
37 hours). Rectal temperature also was significantly different
When all anesthetized groups were analyzed for the among groups 1 through 3 during time period 14-25
effect of treatment on rectal temperature during the hours (P ⫽ .0005). Temperatures were higher in group
early postsurgical period, fentanyl patch presence 2 (mean ⫽ 102.8°F) than group 3 (mean ⫽ 101.2°F),
(P ⫽ .0283) and the application of surgery with the P ⫽ .0001; and lower in group 2 than group 1
fentanyl patch in place (P ⫽ .0314) were both found to (mean ⫽ 103.6°F), P ⫽ .0021.
be significant factors. A higher mean temperature was No significant treatment effects were demonstrated

Fig 2. Mean serum cortisol concentrations for anesthetized Fig 4. Mean rectal temperatures for anesthetized groups
groups (2-5) for time 8-37 hours, including surgical (12-13.5 (2-5) for time 8-37 hours, including surgical (12-13.5 hours)
hours) and early postsurgical (14-25 hours) periods. and early postsurgical (14-25 hours) periods.
356 TRANSDERMAL FENTANYL PATCH
by changes in respiratory rate, heart rate, or systolic
blood pressure.
Evaluation of mean behavior scores revealed pain
scores to be significantly affected by treatment during
the early postsurgical period when analyzing groups 2
through 5 (P ⫽ .0128). The occurrence of surgery
resulted in higher mean pain scores for groups 4 and 5
(1.32 and 1.67, respectively) than groups 2 and 3 (1.13
and 1.15, respectively) (P ⫽ .0089). Pain scores were
not significantly different between groups 4 and 5.
Excitement/sedation scores were not significantly af-
fected by treatment.
Although there was some minor variation in cortisol
concentrations seen over time in group 1, the cortisol
concentrations remained less than 2.9 ␮g/dL (which is
within the normal reference range for cats, 0.5-4.0
␮g/dL) throughout the study and were not significantly
affected by fentanyl concentrations in either the 0- to
37-hour period (P ⫽ .0743) or the entire 0- to 81-hour
time period (P ⫽ .0818) (Fig 5).
DISCUSSION
The transdermal fentanyl patch did influence bio-
chemical markers of perioperative pain and stress.
Cortisol concentrations were significantly lower in the
surgical group that received transdermal fentanyl than
the surgical group that did not in both the surgical and
early postsurgical periods. Because the rise in fentanyl
concentrations in group 1 did not significantly affect
cortisol concentrations, it is presumed that the attenu-
ated rise in cortisol concentrations in group 4 (as
Fig 5. Mean serum cortisol and mean plasma fentanyl con-
centrations for group 1 during the entire study period. (normal
cortisol reference range 0.5-4.0 ␮g/dL).
compared with group 5) resulted from diminished pain
or stress in these cats. These findings are consistent
with the results of a previous opioid efficacy study
evaluating cats undergoing ovariohysterectomy in a
controlled setting.19 However, the present study did
not identify an association between rising cortisol
concentrations and changes in measured clinical vari-
ables. A similar phenomenon was documented in a
clinical study evaluating cats undergoing ovariohys-
terectomy.20 Though surgery generally caused eleva-
tions in blood glucose concentrations in the early
postsurgical period, this effect was dampened in cats
with fentanyl patches. This may be a reflection of the
lower cortisol concentrations in group 4 than group 5
during the early postsurgical period.
Rectal temperature was predictably lowered by
anesthesia. However, a synergistic effect was noted
between the presence of the transdermal fentanyl patch
and anesthesia, resulting in the maintenance of higher
rectal temperatures in cats that underwent anesthesia
with the patch in place. This interaction was overrid-
den by surgery. Opioid agonists are known to cause
hyperthermia in cats following both central and pe-
ripheral administration.21,22
Significant differences were not noted among sur-
gical treatment groups in any parameter measured
beyond 12 hours postoperatively. This fact may be
related to the nature of the pain caused by ovariohys-
terectomy. While ovariohysterectomy undoubtedly
causes pain, perhaps the visceral and superficial so-
matic pain was not intense or prolonged enough to
cause longer-lasting elevations in cortisol concentra-
tions in the surgical group that did not receive anal-
gesics. Visceral pain is generally diffuse and dull,
whereas somatic pain is more localized and sharp.23
Pain associated with an ovariohysterectomy involves
both components, but nonvisceral tissue manipulation
is minimal and superficial. Perhaps an orthopedic
model would cause more profound and prolonged
pain. However, withholding analgesics to establish an
untreated control group in an orthopedic model would
violate accepted standards of care. An alternative
explanation for the lack of a prolonged difference in
cortisol concentrations between surgical treatment
groups may be the physiologic nature of the hormone
itself. Perhaps cortisol is only a reliable marker for
more acute pain due to the presence of cortisol-
mediated negative-feedback mechanisms to both the
hypothalamus and pituitary gland.24
The pain scoring system employed in this study
 GLERUM ET AL
demonstrated differences between operated and non-
operated cats (groups 4 and 5 vs groups 2 and 3), but
it may not have been sensitive enough to detect subtle
differences between operated cats treated with fenta-
nyl and those not treated with fentanyl (groups 4 and
5, respectively).18,25 This is further evidence of the
difficulty in using observational methods to determine
when cats need analgesic therapy. Various scoring
systems have been constructed for the dog and cat,
incorporating behavioral parameters only or combin-
ing both physiologic and behavioral criteria. No single
system has proven to be consistently reliable. Behav-
ioral scores may not correlate well with physiologic
parameters.17,26 Multiple observers may not assess the
same patient in a similar fashion.18 Given behaviors
may even represent different functions among differ-
ent species.26 The simple descriptive scales used in
this study imply equal gradations between numerical
assignments, though this is not necessarily so. How-
ever, the explicit descriptions associated with each
numerical score likely made this scale more conducive
to use by multiple observers. Thoracic wraps were
placed on all individuals throughout the study to
conceal patch presence or absence, but logistics pre-
cluded full blinding of data collectors. Interobserver
variability was minimized by the use of only 2 scorers.
Observer bias also may have led to the assignment of
higher pain scores postoperatively, because observers
could not be blinded to the presence of surgical
treatment.
The transdermal fentanyl patch appears to be a
convenient and effective method for diminishing the
perioperative pain and stress caused by ovariohyster-
ectomy in the cat, as evidenced by the attenuation of
cortisol and blood glucose concentration elevations in
operated cats. Though surgical pain and stress are not
necessarily equivalent,4,27 they are undeniably inter-
related. The transdermal fentanyl patch appeared to be
safe to use in this study and may be effective for
pre-emptive analgesia for visceral and superficial so-
matic pain. Further study is required to determine its
efficacy in treating more intense pain resulting from
more invasive procedures such as orthopedic surgery.
ACKNOWLEDGMENT
The authors thank Dr. Fred Thompson, Kenneth Smith,
and David Smith for their assistance with this project.
357
REFERENCES
1. Johnson JM: The veterinarian’s responsibility: Assessing and
managing acute pain in dogs and cats. Part I. Comp Cont Ed
Pract Vet 13:804-807, 1991
2. Tranquilli WJ, Raffe MR: Understanding pain and analgesic
therapy in pets. Vet Med 84:680-686, 1989
3. Crane SW: Perioperative analgesia: A surgeon’s perspective.
J Am Vet Med Assoc 191:1254-1257, 1987
4. Sanford J, Ewbank R, Molony V, et al: Guidelines for the
recognition and assessment of pain in animals. Vet Rec
118:334-338, 1986
5. Hansen B, Hardie E: Prescription and use of analgesics in
dogs and cats in a veterinary teaching hospital: 258 cases
(1983-1989). J Am Vet Med Assoc 202:1485-1494, 1993
6. Papich MG: Analgesic drug therapy. Semin Vet Med Surg
(Small Animal) 12:80-93, 1997
7. Benson GJ, Thurmon JC: Species difference as a consider-
ation in alleviation of animal pain and distress. J Am Vet
Med Assoc 191:1227-1230, 1987
8. Schultheiss PJ, Morse BC, Baker WH: Evaluation of a
transdermal fentanyl system in the dog. Contemp Top Vet
Med 34:75-81, 1995
9. Kyles AE, Papich M, Hardie EM: Disposition of transder-
mally administered fentanyl in dogs. Am J Vet Res 57:715-
719, 1996
10. Egger CM, Duke T, Archer J, et al: Comparison of plasma
fentanyl concentrations by using three transdermal fentanyl
patch sizes in dogs. Vet Surg 27:159-166, 1998
11. Kyles AE, Hardie EM, Hansen BD, et al: Comparison of
transdermal fentanyl and intramuscular oxymorphone on
post-operative behaviour after ovariohysterectomy in dogs.
Res Vet Sci 65:245-251, 1998
12. Robinson TM, Kruse-Elliott KT, Markel MD, et al: A com-
parison of transdermal fentanyl versus epidural morphine
for analgesia in dogs undergoing major orthopedic surgery.
J Am Anim Hosp Assoc 35:95-100, 1999
13. Moore K, Bjorling D, Manley P, et al: Comparison of
transdermal fentanyl and epidural morphine for analgesia
after triple pelvic osteotomy in dogs. Abstract in Pro-
ceedings of the Ninth Annual ACVS Symposium, 1999,
p 15
14. Scherk-Nixon M: A study of the use of a transdermal fentanyl
patch in cats. J Am Anim Hosp Assoc 32:19-24, 1996
15. Lee DD, Papich MG, Hardie EM: Comparison of pharmaco-
kinetics of fentanyl after intravenous and transdermal ad-
ministration in cats. Am J Vet Res 61:672-677, 2000
16. Yackey M, Ilkiw JE, Pascoe PJ, et al: The effect of transder-
mally administered fentanyl on the minimum alveolar con-
centration of isoflurane in cats. Abstract in Proceedings of
the Sixth International Congress of Veterinary Anaesthesi-
ology, 1997, p 104
17. Holton LL, Scott EM, Nolan AM, et al: Relationship between
physiological factors and clinical pain in dogs scored using
a numerical rating scale. J Small Anim Pract 39:469-474,
1998
358 TRANSDERMAL FENTANYL PATCH
18. Firth AM, Haldane SL: Development of a scale to evaluate
postoperative pain in dogs. J Am Vet Med Assoc 214:651-
659, 1999
19. Smith JD, Allen SW, Quandt JE, et al: Indicators of postop-
erative pain in cats and correlation with clinical criteria.
Am J Vet Res 57:1674-1678, 1996
20. Smith JD, Allen SW, Quandt JE: Changes in cortisol concen-
tration in response to stress and postoperative pain in
client-owned cats and correlation with objective clinical
variables. Am J Vet Res 60:432-436, 1999
21. Clark WG: Effects of opioid peptides on thermoregulation.
Fed Proc 40:2754-2759, 1981
22. Clark WG, Cumby HR: Hyperthermic responses to central and
peripheral injections of morphine sulphate in the cat. Br J
Pharmacol 63:65-71, 1978
23. Guyton AC, Hall JE: Somatic sensations: II. Pain, headache,
and thermal sensations, in Textbook of Medical Physiology.
Philadelphia, PA, Saunders, 1996, pp 609-620
24. Guyton AC, Hall JE: The adrenocortical hormones, in Text-
book of Medical Physiology. Philadelphia, PA, Saunders,
1996, pp 957-970
25. Holton LL, Scott EM, Nolan AM, et al: Comparison of three
methods used for assessment of pain in dogs. J Am Vet Med
Assoc 212:61-66, 1998
26. Conzemius MG, Hill CM, Sammarco JL, et al: Correlation
between subjective and objective measures used to deter-
mine severity of postoperative pain in dogs. J Am Vet Med
Assoc 210:1619-1622, 1997
27. Kehlet H: Surgical stress: The role of pain and analgesia. Br J
Anaesth 63:189-195, 1989