Recognizing Pain and Distress in Laboratory Animals

E. Carstens and Gary P. Moberg

Introduction

The recognition of pain, stress, and distress is critical in

maintaining the well-being of laboratory animals. However,
this important task is difficult because of a lack of
agreed-upon definitions of these terms, as well as the absence of absolute, objective measures. Because animals cannot verbalize what they are experiencing, investigators and
caretakers must deduce the animal's condition based on appearance and behavior. Excellent sources of information on
these issues have appeared previously (Bateson 1991; CCAC
1998; IRAC 1985; Morton and Griffiths 1985; NRC 1992;
Short and Van Poznak 1992; Smith and Boyd 1991; Soma
1987; UFAW 1989; Wallace et al. 1990). In this article, we
discuss the biologic meaning of pain, stress, and distress,
with the aid of a model, and review and update guidelines for
the recognition and assessment of pain and distress in laboratory animals.

Pain
The International Association for the Study of Pain has defined
pain in humans as "an unpleasant sensory and emotional
E. Carstens, Ph.D., is Professor in the Section of Neurobiology, Physiology
and Behavior, University of California, Davis. Gary P. Moberg, Ph.D., now
deceased, was Professor in the Department of Animal Science, University
of California, Davis, when this manuscript was submitted.

62

Acute pain normally warns of the presence of a potentially injurious stimulus and has great adaptive value. Delivery
of a noxious stimulus to an extremity usually elicits a protective, stereotyped reflexive withdrawal of the limb away from
the source of injury. This may or may not be associated with
other nocifensive reactions such as vocalization, orientation
toward the stimulus, escape or attack, shaking, rubbing,
scratching, biting or licking of the stimulated area, and autonomic reactions such as increased blood pressure and heartrate, piloerection, and pupillary dilation.
Considerable evidence indicates that simple nocifensive
behavioral reactions, such as limb withdrawal reflexes, are
generally elicited at similar thresholds in animals and humans
and that the magnitude of the response increases with stimulus
intensity. The presence of other more complex, integrated
behavioral reactions varies across and within species and
even within individuals. The maximal stimulus intensity voluntarily accepted by a human is the tolerance limit. The pain
tolerance limit in animals, assessed subjectively, varies
widely across species with some (e.g., cattle) apparently
being more "stoic." The relatively few studies measuring the
magnitude of nocifensive behavioral responses (e.g., operantly conditioned escape responses in monkeys) suggest that
at least some species perceive increasing pain over the same
ILAR Journal

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

e have long recognized the impact that pain, stress,

and distress can have on the welfare of laboratory
animals, or for that matter, any animal. Pain and
stress can cause distress, and the biologic effects can compromise experimental results (Moberg 1999). For these reasons, we are continually striving to control and ameliorate
the effects of pain and stress on laboratory animals. Neither
pain nor stressa part of an animal's life, as they are ours
can ever be totally eliminated. Pain normally serves as a
protective function to warn of impending danger and is therefore adaptive. Animals have evolved appropriate biologic
strategies to assist them in coping with stress. Thus, pain and
stress are not inherently bad for an animal, unless these biologic strategies fail to protect the animal from stress or the
biologic cost of coping takes too great a toll on the animal.
Then the animal experiences distress and its welfare is threatened. The question is, when does an animal cross over from
nonthreatening stress to distress?

experience associated with potential or actual tissue damage,

or described in such terms." In the absence of language, it is
impossible to know an animal's experience, which can be
deduced only by observing the animal's general appearance
and pain-like ("nocifensive") behavior. Thus, most mammalian species react in a manner to avoid and reduce the impact
of acute noxious (potentially or overtly tissue-damaging)
stimuli, suggesting that they perceive pain. Considerable
evidence indicates that common laboratory animals possess
nervous systems that receive and process noxious stimuli in
a manner similar to humans, at least at subcortical levels.
Although it may be safe to assume that higher mammals
consciously perceive pain, it has been argued that because
their cerebral cortices are smaller and less developed than
humans', they may not have the capacity to comprehend the
meaning of pain or an awareness of impending doom (Bateson
1991;Carli 1980; Melzack and Dennis 1980; Smith and Boyd
1991). Thus, it is arguable whether laboratory and companion animals suffer in the same sense as humans, but we believe that animals should be treated as if they do, recognizing
that there are no objective or measurable indices of suffering.
In the following sections, we attempt to identify indicators of
pain, stress, and distress in mammalian laboratory species.

After injury to peripheral or central nervous tissue, or

progressive disease states such as cancer, pain may develop
over time and persist indefinitely. This chronic pain no longer
has adaptive value but instead causes distress. Although more
difficult to recognize than acute pain, it is crucial that signs
of chronic pain are identified to afford appropriate treatment
and minimize or prevent distress.

Neural Basis of Pain

Acute pain sensation is normally elicited by noxious stimulation of high-threshold receptors, called nociceptors, which
exist in most tissues. Varieties of nociceptors exist that respond selectively to intense mechanical stimuli (mechanical
nociceptor), or nonselectively, to noxious mechanical, thermal, and chemical stimuli ("polymodal" nociceptor). Some
nociceptors do not respond to noxious stimuli initially but
begin to respond during development of inflammation
("sleeping" nociceptors). The afferent fibers of nociceptors
travel in peripheral nerves into the spinal cord, where they
synaptically contact second-order neurons in the dorsal horn.
Most nociceptors are thought to release the neurotransmitter
glutamate, as well as the neuroactive peptides substance P
and neurokinin A, from their presynaptic terminals. These
transmitters excite the second-order dorsal horn neurons,
which either send axons into ascending sensory pathways to
transmit pain information to higher centers or serve as interneurons in segmental reflex pathways. Ascending pathways
involved in the transmission of pain include the spinothalamic tract, the spinocervicothalamic tract, the spinohypothalamic tract, spinoreticular pathways, the spinoparabrachial
amygdalar pathway, the postsynaptic dorsal column pathway, and a recently identified pathway for visceral pain in
the dorsal columns. In addition, pain may be transmitted
rostrally by multisynaptic connections (for recent reviews,
Volume 4 1 , Number 2

2000

see Millan 1999; Willis 1985; Willis and Coggeshall 1991;

Willis and Westlund 1997).
After a peripheral injury such as surgical incision, the
resultant hyperalgesia has two possible explanations. First, it
is known that noxious stimulation can cause nociceptors to
"sensitize," that is, to give a greater response to subsequent
stimuli. This action is due to the release of algesic chemicals
from damaged cutaneous cells into the extracellular environment, where they can depolarize the fiber endings of nociceptors. In addition, injury can lead to sensitization of
second-order spinal neurons, involving a multistep process
starting with the release of an excessive amount of glutamate
from the nociceptor presynaptic terminals. As a result of the
dual action of glutamate to depolarize the dorsal horn neuron, and to open cation channels by its action at a subtype of
postsynaptic glutamate receptor called the NMDA-receptor,
Ca++ gains entry into the postsynaptic neuron to trigger second messenger systems ultimately leading to increased excitability of the neuron (Dickenson et al. 1997). This process,
called central sensitization, can lead to hyperalgesia because
the hyperexcitable neuron now sends a larger pain signal to
higher centers. The neuron can even be excited by a nonnoxious tactile stimulus; the heightened signal sent to the
brain is interpreted as painful even though it was caused by a
normally nonpainful stimulus. Both peripheral and central
sensitization very likely contribute to increased pain sensitivity after peripheral injuries (for recent reviews, see
Carstens 1995; Dubner and Ruda 1992; Millan 1999).
Pain-transmitting spinal neurons are known to be under
powerful descending modulation from the brainstem. It
should be noted that descending pathways can either inhibit
or excite spinal neurons (Fields and Basbaum 1994; Maier et
al. 1992; Millan 1999; Sandkuhler 1996; Willis 1985). These
two forms of descending modulation probably play important roles in the complex effects that environmental stressors
can have in increasing or reducing pain sensitivity.
The neural mechanisms underlying the development of
chronic pain are not as well understood and are currently
under active investigation. Chronic pain likely involves central sensitization (see above) and other long-term alterations
in the function of pain-signaling pathways, such as remodeling of synapses within the spinal cord (Lekan et al. 1996;
Millan 1999).

Recognition of Pain in Laboratory Animals

It may not always be obvious that a laboratory animal is in
pain, and careful observation of changes in appearance and
behavior over time will aid in recognizing signs of pain.
Knowledge of the animal's treatment or procedure will obviously assist in this assessment. It is advisable to assume that
"unless the contrary is established, investigators should consider that procedures that cause pain or distress in human
beings may cause pain or distress in other animals" (IRAC
1985). A very general set of observations for assessing pain
and distress has been proposed (Morton and Griffiths 1985):
63

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

range of stimulus intensities (from threshold to tolerance

limit) as humans (Cooper and Vierck 1986a).
Consequent to surgery, accidental injury, illness, or disease, a persistent state of pain may develop. After an injury
such as a surgical incision, tissue at and surrounding the site
of trauma may exhibit increased pain sensitivity. This sensitivity is usually characterized by an increase in the intensity
of pain (i.e., nocifensive response) elicited by a noxious
stimulus, termed hyperalgesia. This condition is frequently
accompanied by allodynia, in which pain is elicited by a
nonnoxious stimulus (e.g., touch) that normally is incapable
of eliciting pain. Hyperalgesia and allodynia are adaptive in
that the animal is self-motivated to reduce movement to minimize reinjury and aid recuperation, which normally requires
a few days. Although hyperalgesia and allodynia can be objectively measured experimentally (e.g., Bennett and Xie
1988; Tabo et al. 1999), these and other indicators of pain are
difficult to quantify objectively in the clinic. The use of scaling procedures by veterinarians to assess an animal's degree
of postsurgical pain may prove to be helpful (e.g., Holton et
al. 1998).

itch, which is very difficult to differentiate from pain. These

changes may occur in the absence of more general alterations in appearance, feeding, and other behaviors. After a
more severe injury such as transection of a peripheral nerve
or dorsal roots, the animal may self-mutilate the denervated
limb; however, it is not yet known whether this behavior
represents a reaction to pain, a response to a nonpainful abnormal sensation such as tingling, or failure of the animal to
recognize the denervated limb as part of its body schema
(Coderre et al. 1986).
Signs of chronic pain due to pathology may be more
difficult to recognize initially because they are likely to develop slowly and vary in magnitude over time. Indicators of
pain include a general deviation from the animal's normal
healthy appearance due to reduced grooming, loss in body
weight, and a reduced locomotor activity (Table 1). There
may be postural changes, such as a "hunched" posture with
abdominal pain. Rats may exhibit porphyrin secretions ("red

Table 1 Indicators of pain in several common laboratory animals'

Species

General behavior

Appearance

Physiology

Rat

Reduced activity; reduced appetite; reduced

drinking; licks; guards limbs; self-mutilates;
increased aggression and vocalization; aversion toward con-specifics

Ungroomed; piloerection; abnormal

stance; hunched posture ("dormouse"); "red tears"; eyelids partly
closed; pupils dilated; nasal discharge; recumbent

Sleep disrupted;
hypothermia; rapid
shallow breathing,
may grunt on expiration

Mouse

Similar to rat; increased vibrassal movement

Similar to rat; no ocular porphyrin

secretion

Similar to rat

Guinea pig

Squeals, stampedes when handled; or quiet

Similar to rat

Similar to rat

Rabbit

Anxious; hides; squeals or cries; or aggressive; scratches/bites; reduced appetite; cannibalizes young; tonic immobility

May not show large change

Salivates; rapid
shallow breathing

Dog

Bites, scratches, guards; whimpers or howls;

growls; quiet, submissive; or more aggressive
to handling

Stiff, moves less or lies still; "hangdog" look; tail between legs

Shivers; pants; urinates

Cat

Quiet; hisses or spits; hides; licks excessively;

limps; guards limbs; stilted gait; reduced
appetite; escape response to handling

Apprehensive facial expression;

tucks-in limbs; hunches head/neck;
ungroomed; flattens ears; cringes

Horse

Reluctant to be handled; decreased activity;

disrupted feeding; restlessness or depression;
reluctance to move; unusual limb position

Anxious appearance; dilated pupils;

flared nostrils; glassy eyes; lowers
head

Nonhuman
primate

Screams or moans; reduced feeding and

drinking; aggression

Crouches; "sad" facial expression

or grimace; may stop grooming

Sweats

Adapted from Morton DB, Griffiths PHM. 1985. Guidelines on the recognition of pain and discomfort in experimental animals and an hypothesis for assessment. Vet Rec 116:431-436; NRC [National Research Council]. 1992. Recognition and Alleviation of Pain and Distress in
Laboratory Animals. Washington DC: National Academy Press; Soma LR. 1987. Assessment of animal pain in experimental animals. Lab
Anim Sci 37:71-74; and Wallace J, Sanford J, Smith W, Spencer V. 1990. The assessment and control of the severity of scientific procedures
on laboratory animals. Report of the Laboratory Animal Science Association Working Party. Lab Anim 24:97-130.

64

IIAR Journal

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

change in body weight, external physical appearance, clinical signs, changes in unprovoked behavior, and behavioral
responses to external stimuli. These criteria provide a good
framework for additional assessment.
For animals that have been subjected to a localized surgical procedure, one would look for signs of pain in the affected body area. For example, in animal models of pain due
to surgical incision or partial nerve injury, one should observe the affected limb for signs of spontaneous pain and/or
hyperalgesia and allodynia. These signs include altered posture and gait with less weight placed on the injured limb,
limb guarding, change in muscle tone and limb temperature,
vocalization and/or exaggerated withdrawal of the limb to
innocuous or noxious stimulation, and possibly also shaking,
licking, scratching or biting of the affected area (Bennett and
Xie 1988; Kim and Chung 1992; Selzer et al. 1990; Tabo et
al. 1999; Takaishi et al. 1996; reviewed in Millan 1999).
However, scratching of the injured area might also reflect

Influence of Stress on Pain

Although pain is an important stressor, many forms of environmental or psychological stress can also increase or decrease pain perception. Examples of stressors that can reduce
nocifensive behavioral responses in animals, resulting in
so-called "stress-induced analgesia," include inescapable
footshock, centrifugal rotation, noxious stimuli, and presumably more psychological stressors such as fear or aggression
(Amit and Galina 1988; Tricklebank and Curzon 1984;
Watkins and Mayer 1982). Stress-induced analgesia is
thought to be mediated in part by descending pathways from
the brain that can modulate the spinal transmission of pain
signals (see above). Reduction of pain would be advantageous under highly stressful circumstances, such as
fight-or-flight, in which it would be life threatening for an
animal to attend to a painful injury rather than delaying until
after it has reached safety. The capacity for endogenous systems to modulate pain in humans is well documented, a good
example being the soldier who is injured on the battlefield
but perceives no pain until long after he has been removed to
the safe environment of the hospital. Indeed, a substantial
fraction of patients who were admitted to the emergency
room of a major hospital for severe injuries, such as broken
bones, reported that they did not feel any pain for up to 7 hr
after incurring the injury (Melzack et al. 1982). Thus, in
assessing pain in animals, it must be recognized that the
animal's level of pain might well be influenced by the history of other stressors that have impinged on it, as in humans.
In this context, the poorly understood relationship between pain and illness deserves brief mention. Sickness and
fever can be induced in rats by a bacterial lipopolysaccharide
(endotoxin) (Romanovsky et al. 1996; Watkins et al. 1994).
During the initial phase of lipopolysaccharide-induced fever,
rats exhibit hyperalgesia and show increased locomotor
activity, whereas they exhibit hypoalgesia during the later
stage that is characterized by lethargy and poor temperature
regulation. The significance of this biphasic change in pain

Volume 4 1 , Number 2

2000

sensitivity is not clear but certainly suggests that the assessment of pain in animals might be further complicated by the
presence of systemic infection.

Animal Stress and Distress: A Model

It has never been possible to accurately define stress because
the term is applied so broadly. Some authors use stress to
describe the biological response of an animal to a threat.
Others use stress to describe the actual threat, such as restraint stress or temperature stress. For this discussion, we
define stress as the biological responses an animal exhibits in
an attempt to cope with a threat to its homeostasis. The threat
to its homeostasis is the stressor. No stress occurs unless the
animal perceives a threat, either consciously or unconsciously. Although animals are conscious of most stressors,
there are potential threats to an animal's homeostasis, such
as tumor growth, that will elicit stress responses even though
the animal may never be conscious of the threats. For this
reason, we define perception very broadly to include both
conscious and unconscious recognition of a stressor. Regardless of how an animal perceives a stressor, the resulting stress
may or may not prove to be distressful.
To understand how stress leads to distress, we will use
the model of animal stress depicted in Figure 1 to place the
diverse literature on animal stress into a working framework.
Because the development of the model is discussed elsewhere (Moberg 1985, 1999, 2000), we focus herein on the
aspects of the model that are relevant to distress. All stress
responses, good, bad, or inconsequential, result from the animal responding to a stressor by exhibiting one or more of
four general biological defense responses: behavioral, autonomic nervous system, neuroendocrine system, or immune.
The responses of these systems to the stressor alter biological functions, resulting in the stress responses. Because most
stressors are brief, the changes in biological function required
to cope with the stressor are minimal and of no significant
consequence to the animal (Figure 2), that is, the biological
cost of the stress is of little consequence to the animal's
well-being. As schematically depicted in Figure 2, the biological resources required by the stress response do not affect
other biological functions (Moberg 1999, 2000). Once the
stressor is alleviated, the animal returns to prestress conditions.
When the stressor is severe, of long duration, or characterized by the cumulative effects of several stressors, then
the total biological cost of the stress response may require
the diversion of resources from other biological activities
(Figure 3), disrupting other biological functions that are critical to the animal's well-being. When these functions are
disrupted, the animal enters into a prepathological state (Figure 1), rendering it vulnerable to the development of pathologies (Moberg 1999, 2000). During this time, when normal
function is disrupted and the possibility for pathology exists,
the animal experiences distress and its welfare is threatened.
In this model of stress, we view biological resources and
pathology in the broadest sense. Although shifts in energy
65

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

tears") around the eyes. The animal may vocalize spontaneously and/or when handled; however, vocalization is not specific to pain nor do animals in pain necessarily vocalize (Cooper and Vierck 1986b). Rodents emit vocalizations in the
audible range as well as at frequencies >20 KHz; and although such ultrasonic vocalizations can be associated with
pain (Jourdan et al. 1995,1998), they are also emitted under
other nonpainful circumstances. Finally, an animal in pain
may fail to exhibit normal curiosity or social interactions, for
example, withdrawing from the handler or, alternatively,
showing increased aggressiveness. Here, knowledge of
species-specific behavior is important in assessing pain.
Table I provides a summary of many indicators of pain in
several species of laboratory and companion animal. Detailed descriptions of signs of pain in other animal species
are available in the literature (Morton and Griffiths 1985;
NRC 1992; Soma 1987; Wallace et al. 1990).

inhumane and unacceptable. Our goal must be to recognize

distress before the animal reaches the pathological state. Because the key to distress is the biologic cost of stress, we
need to focus on the biologic processes leading to the change
in biologic function or the alteration in function as indications of distress occurring. Because there is currently no litmus test for distress, we need to approach the recognition of
distress on almost a case-by-case basis.

Nonpain Distress
In some aspects, recognizing nonpain distress is more difficult than recognizing the discomfort associated with pain;
although the source of pain is usually obvious, our challenge
is to gauge the level of discomfort. In contrast, many nonpain
stressors confronting the laboratory animal are less obvious
and their biologic effects on the animal are poorly understood. These stressors fall into two general categories: stressors
associated with the experimental manipulation of the animal
and stressors resulting from routine husbandry practices.

STIMULUS

RECOGNITION OF A
THREAT TO
HOMEOSTASIS

CENTRAL NERVOUS SYSTEM

PERCEPTION OF STRESSOR

ORGANIZATION OF BIOLOGICAL
DEFENSE

BIOLOGICAL RESPONSE
(Behavioral, Autonomic,
Neuroendocrine, Immunological)

STRESS RESPONSE
NORMAL
BIOLOGICAL FUNCTION
r

ALTERED
BIOLOGICAL FUNCTION
CONSEQUENCES OF
STRESS

PREPATHOLOGICAL STATE
DEVELOPMENT OF PATHOLOGY

Figure 1 Model of the biologic response of animal to stress. Reprinted with permission from Moberg GP. 1999. When does stress become
distress? Lab Anim 28:422-426.
66

ILAR Journal

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

utilization characterize many stresses (Elsasser et al. 2000;

Moberg 2000), other changes in function, such as altered
behavior or loss of immune competence, are biological resources effected in an attempt to cope with the stressor. Likewise, disease is not the only type of pathology an animal
experiences during stress (Moberg 1985). Loss of reproduction, abnormal behavior, or failure to grow are examples of
stress-induced pathologies. The presence of any of these pathologies or the biologic conditions leading to these pathologies (prepathological state) clearly indicates that the animal
is in distress and its well-being is threatened (Moberg 1985,
1999,2000). Although it is difficult to objectively assess the
degree to which an animal's well-being is compromised, a
list of some obvious indicators is provided in Table 2 (NRC
1992).
Although providing a biologic definition of distress is
not difficult, recognizing when distress occurs is far more
difficult. Certainly the presence of pathologies such as disease, selfmutilation, or death are obvious indicators of distress; however, waiting for the occurrence of such endpoints
as an indication of distress occurring in laboratory animals is

STRESS RESPONSE
STRESS

(0
0
O
3
O
(0

F1

F1

F2

F2

F2

F3

F3

F3

F"n"

F"n'

BASAL
FUNCTIONS

BASAL
FUNCTIONS

Normal
Function

Stress

o
tL

75
o
"5>
o
o
ffi

Recovery

F"n"

BASAL
FUNCTIONS

Normal
Function

Figure 2 Hypothetical scheme of how stress diverts biologic resources during mild stress. In this scheme, biologic resources are arbitrarily
assigned to various biologic functions (Fl-F"n"). During mild stress, only reserve resources are used to cope with the stressor. The total stress
response extends from the time biologic resources are diverted until the reserves have been replenished. Reprinted with permission from
Moberg GP. 1999. When does stress become distress? Lab Anim 28:422-426.

Both categories share the same biology of the stress response,

but recognition of these stressors requires different approaches.
Distress does not occur without stress. As discussed, distress occurs only when the biological cost of stress negatively affects biological functions critical to the animal's
well-being. Stress, however, has proved difficult to measure
(Moberg 1985). Returning to the model of animal stress (Figure 1), it would seem that the most reasonable strategy for
measuring stress would be to monitor the responses of the
four major defense systems (behavior, autonomic nervous
system, neuroendocrine system, and immune system) inasmuch as they are responsible for the biological changes that
occur during stress. Indeed, all four systems have been used
to measure stress (Moberg 2000); however, none have proved
to be a reliable measure of stress, let alone distress. No one
system responds to all stressors. Frequently, they respond
comparably to both threatening and nonthreatening stimuli.
In addition, there are intraanimal variations in responses to
the same stressor that are very difficult to monitor outside the
controlled conditions of the laboratory.
Volume 4 1 , Number 2

2000

The stressor responses of the hypothalmic pituitary adrenal axis (HPA1) provide an example of the difficulty encountered in measuring stress. Measuring the secretion of the
HPA hormones, especially the glucocorticosteroids cortisol
and corticosterone, has been the most popular tool for evaluating stress, and researchers frequently use increases in circulating glucocortico steroids as proof of stress. Certainly
numerous stressors do elicit an increase in circulating steroids but contrary to Selye's (1950) prediction, not all stressors elicit an HPA response (Mason 1968). Further complicating the use of the HPA axis as a measure of stress is the
comparable response of the axis to both potentially threatening and nonthreatening stimuli. This is perhaps best exemplified by studies of Colborn and colleagues (1991) who found
that stallions secreted similar amounts of cortisol whether
the horses were restrained, exercised, or allowed to mate. It
is difficult to argue that mating has the same negative impact
on the stallion's welfare as being restrained.
'Abbreviation used in this article: HPA, hypothalmic pituitary adrenal axis.

67

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

F1

DISTRESS

(0
F1

3
O
(0

F2

F1

STRESS

F3

F3
F1
F2
F3

F"n"

F"n"
F"n"

BASAL
FUNCTIONS

BASAL
FUNCTIONS

Normal
Function

Stress

BASAL
FUNCTIONS

Normal
Function

Figure 3 Hypothetical scheme of how the diversion of biologic resources necessary to cope with severe stress significantly impacts other
biologic functions leading to distress. As compared with mild stress (Figure 2), the biologic cost of distress requires a much longer recovery
period. Reprinted with permission from Moberg GP. 1999. When does stress become distress? Lab Anim 28:422-426.

Further complicating stress measurements are the intraanimal differences in how the four general defense systems
respond in attempting to cope with the stressor (Engle 1967;
Henry 1992; Moberg 1985,2000; Weiss 1972). Early experience, genetics, age, and physiological state are examples of a
multitude of modulators that influence the nature of the stress
response (Moberg 1985, 2000). With traditional laboratory
animals such as rodents, many of these variables can be controlled and accounted for in the experimental design; however, for some laboratory animals (e.g., nonhuman primates
or random source animals), it is extremely difficult to account for these modulators of the stress response because
simple measures of hormones, autonomic nervous system
activity, or immune responses may be unreliable measures of
stress outside the experimental paradigm (Moberg 1987b).
In monitoring the responses of the neuroendocrine, autonomic, and immune systems to potential stressors, it is very
difficult to obtain the requisite samples for analysis without
allowing the sampling procedures to stress the animals. To
evaluate these systems, it is usually necessary to obtain blood
samples or attach monitoring equipment. The act of restrain68

ing animals and drawing blood is a stressor for the animal,

resulting in significant changes in circulating HPA hormones
as well as changes in other neuroendocrine hormones or autonomic nervous system activity. Sometimes it is possible to
incorporate sampling methods into the experimental design,
permitting the investigator to obtain samples via cannulae or
some other methods without further stressing the animals;
however, this approach is feasible for only a limited number
of animals and is certainly impractical for monitoring distress related to the husbandry of laboratory colonies, especially for such species as nonhuman primates.
Of the four major defense systems, behavior is perhaps
the most promising as an unobtrusive way to monitor for
stress (Mench 1998). Although promising, the use of behavior as a measure of stress is hampered by our lack of
understanding of animal behavior as it relates to stress
(Rushen 2000). Accepting behavior as an indicator of distress requires the correlation of reliable behavioral changes
with stress-induced biologic changes that force the animal
into the prepathological state.
Even if an animal experiences stress, it does not mean
WAR Journal

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

"J5
o
"5>
o
o

F2

Table 2 Some behavioral, physiologic, and biochemical

indicators of well-being*'*
Behavioral

Physiological

Biochemical

Grooming
Appetite
Activity
Aggression
Facial expression
Vocalization
Appearance
Posture
Response to handling

Temperature
Pulse
Respiration
Weight loss
Blood-cell count
Blood-cell structure
Cardiac output
Blood flow

Corticosteroids
Catecholamines
Thyroxin
Prolactin
P-Endorphin
ACTH
Glucagon
Insulin
Vasopressin
Substance P

that the animal suffers distress. As we have stated, stress is a

part of life. If the biologic cost of the stress has no impact on
other biologic functions (Figure 2), then there is no distress.
However, if the biologic cost of the stress affects other biologic functions, distress does occur (Figure 3). From this
argument, it would appear that the most appropriate measures of distress would be changes in biologic functions (see
Figure 1) that lead to the prepathological and pathological
states (Moberg 1996). Although morbidity and mortality are
not acceptable for monitoring distress, their presence must
not be ignored as indicative of severe distress occurring in
the colony. Likewise, failure to grow, reduction in reproductive success, and markedly abnormal behavior are indications of distress. Nevertheless, we need methods to determine when the prepathological state is developing. Reduction
in immune competence (Blecha 2000), disruption of the reproductive neuroendocrine axis (Moberg 1987a), altered
metabolism (Elsasser et al. 2000), and growth (Moberg 1999)
have been used to identify distress; however, use of these
measures is as tedious and difficult as measuring hormone
secretion or autonomic nervous system activity. The advantage of evaluating these biologic functions for detecting distress, as opposed to simply measuring hormone secretion, is
that we know the stressor has disrupted biologic functions
that truly affect the animal's well-beinghealth, reproduction, and growth.
Although difficult, taking such measures of stress and
distress can be incorporated into the design of many experiments. However, using them to identify distress in animal
colonies resulting from improper husbandry practices is far
more difficult because of the sheer number and types of animals involved. Modem animal colonies, through controlled
environments and standardized management practices, do
not experience many potential stressors such as extreme temVolume 4 1 , Number 2

2000

peratures, inadequate nutrition, or abnormal lighting schedules; however, many other subtle environmental factors may
cause distress. In their superb report on the recognition and
alleviation of pain and distress in laboratory animals, the
National Research Council (NRC 1992) identifies such potential factors as the animal's relationships with conspecifics, the necessary available social space, the potential effects
of feeding and foraging behavior, and the physical nature of
the captive environment. Unfortunately, apart from studies
of environmental enrichment in certain primate species, not
enough research on these topics has emerged since this report was issued to guide us in deciding how to house our
animals to reduce or even evaluate potential distress related
to these concerns.
In the absence of simple, definitive measures of distress,
our best approach is to use our intuition and sensitivity to
reduce potential distress in laboratory colonies. Because the
presence of distress is related to the biologic cost of stress,
the simplest approach is to reduce the burden of this cost.
Although a single stressor may induce a stress response with
a sufficient biologic cost to induce distress, it is probably
more common for distress related to husbandry to result from
the biologic costs of several small stressors accumulating to
result in a total cost of stress that is sufficient to induce
distress (Moberg 1999, 2000). Recognizing this cumulative
effect is especially important in developing management
practices for laboratory animals. Any one manipulation, such
as transfer to a new cage, might be inconsequential to die
animal. However, if several such practices are combined
within a time frame when the individual costs can accumulate, the animal may experience distress, its welfare may be
jeopardized, and the distress stemming from routine husbandry of the animal might confound experimental results.
By the same token, what might be viewed as innocuous
69

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

a
Departures from normal behaviors and characteristics are suggestive of
changes in well-being. A knowledge of species-typical and individual-specific
behaviors and clinical values is essential.
b
Reprinted from NRC [National Research Council]. 1992. Recognition and
Alleviation of Pain and Distress in Laboratory Animals. Washington DC:
National Academy Press, with permission.

manipulation of the animal may confound experimental results. For example, a growing mouse restrained for a single
4-hr period will require 48 hr to return to normal body weight
and both protein and lipid energy (K. D. Laugero, University
of California, San Francisco, and G. P. Moberg, 1998, personal communication). By being sensitive to the potential
biologic costs of individual stressors, it is possible to develop
husbandry and experimental protocols that eliminate or ameliorate distress, even in the absence of definitive measures of
stress or distress.

References

70

ILAR Journal

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

Amit Z, Galina ZH. 1988. Stress induced analgesia plays an adaptive role in
the organization of behavioral responding. Brain Res Bull 21:955-958.
Bateson P. 1991. Assessment of pain in animals. Anim Behav 42:827-839.
Bennett GJ, Xie YK 1988. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 3:87-107.
Blecha F. 2000. Immune system response to stress. In: Moberg GP, Mench
JA, eds. Biology of Animal Stress: Implications for Animal Welfare.
Wallingford, Oxon, UK: CAB (Forthcoming).
Carli G. 1980. The existence of pain in animals. In: Kosterlitz HW, Terenius
LY, eds. Pain and Society. Weinheim: Verlag Chemie. p 27-36.
Carstens E. 1995. Neural mechanisms of hyperalgesia: Peripheral or central
sensitization? News Physiol Res 10:260-265.
CCAC [Canadian Council on Animal Care]. 1998. Guidelines on choosing
an appropriate endpoint in experiments using animals for research, teaching and testing. Ottawa Ontario Canada: CCAC. (Also available at
<http ://w w w.ccac .ca/english/gdllnes/endpts/appopen.htm>).
Coderre TJ, Grimes RW, Melzack R. 1986. Deafferentation and chronic
pain in animals: An evaluation of evidence suggesting autotomy is related to pain. Pain 26:61-84.
Colborn DR, Thompson DL Jr, Roth TL, Capehart JS, White KL. 1991.
Responses of cortisol and prolactin to sexual excitement and stress in
stallions and geldings. J Anim Sci 69:2556-2562.
Cooper BY, Vierck CJ Jr. 1986a. Measurement of pain and morphine hypalgesia in monkeys. Pain 26:361-92.
Cooper BY, Vierck CJ Jr. 1986b. Vocalizations as measures of pain in
monkeys. Pain 26:393-407.
Dickenson AH, Chapman V, Green GM. 1997. The pharmacology of excitatory and inhibitory amino acid-mediated events in the transmission
and modulation of pain in the spinal cord. Gen Pharmacol 28:633-638.
Dubner R, Ruda MA. 1992. Activity-dependent neuronal plasticity following tissue injury and inflammation. Trends Neurosci 15:96-103.
Elsasser TH, Klasing KC, Filipov N, Thompson F. 2000. The metabolic
consequences of stress: Targets for stress and priorities of nutrient use.
In: Moberg GP, Mench JA, eds. Biology of Animal Stress: Implications
for Animal Welfare. Wallingford, Oxon, UK: CAB Int (Forthcoming).
Engle GI. 1967. A psychological setting of somatic disease: The "giving
up-given up" complex. Proc R Soc Med 60:553-555.
Fields HL, Basbaum Al. 1994. Central nervous system mechanisms of pain
modulation. In: Wall PD, Melzack R, eds. Textbook of Pain, 3rd ed.
Edinburgh: Churchill-Livingstone, p 243-257.
Henry JP. 1992. Biological basis of the stress response. Integrat Physiol
Behav Sci 27:66-83.
Holton LL, Scott EM, Nolan AM, Reid J, Welsh E, Flaherty D. 1998.
Comparison of three methods used for assessment of pain in dogs. J Am
Vet Med Assoc 212:61-66.
IRAC [Interagency Research Animal Committee]. 1985. US Government
Principles for Utilization and Care of Vertebrate Animals Used in Testing, Research and Training. Prepared by the Interagency Research Animal Committee, National Institutes of Health (Lead), p 81-83 in Guide
for the Care and Use of Laboratory Animals. (NTH publication no. 85-23).
Washington DC: US Department of Health and Human Resources.

Jourdan D, Ardid D, Chapuy E, Eschalier A, Le Bars D. 1995. Audible and

ultrasonic vocalization elicited by single electrical nociceptive stimuli to
the tail in the rat. Pain 63:237-249.
Jourdan D, Ardid D, Chapuy E, Le Bars D, Eschalier A. 1998. Effect of
analgesics on audible and ultrasonic pain-induced vocalization in the
rat. Life Sci 63:1761-1768.
Kim SH, Chung J-M. 1992. An experimental model for peripheral neuropathy
produced by segmental spinal nerve ligation in the rat. Pain 50:355-363.
Lekan HA, Carlton SM, Coggeshall RE. 1996. Sprouting of A-beta fibres
into lamina II of the rat dorsal horn in peripheral neuropathy. Neurosci
Lett 208:147-150.
Maier SF, Wiertelak EP, Watkins LR. 1992. Endogenous pain facilitory
systems. Am Pain Soc J 1:191-198.
Mason JW. 1968. "Over-all" hormonal balance as a key to endocrine organization. Psychosomat Med 30:791-808.
Melzack R, Dennis SG. 1980. Phylogenetic evolution of pain expression in
animals. In: Kosterlitz HW, Terenius LY, eds. Pain and Society. Weinheim: Verlag Chemie. p 13-26.
Melzack R, Wall PD, Ty TC. 1982. Acute pain in an emergency clinic:
Latency of onset and descriptor patterns related to different injuries.
Pain 14:33-43.
Mench JA. 1998. Why it is important to understand animal behavior. ILAR
J 39:20-26.
Millan JJ. 1999. The induction of pain: An integrative review. Prog Neurobiol 57:1-164.
Moberg GP. 1985. Biological response to stress: Key to assessment of animal well-being? In: Moberg GP, ed. Animal Stress. Bethesda MD:
American Physiological Society.
Moberg GP. 1987a. Influence of the adrenal axis upon the gonads. In:
Clarke J, ed. Reviews of Reproductive Biology 9:456-496.
Moberg GP. 1987b. Problems in defining stress and distress in animals. J
Am Vet Med Assoc 191:1207-1211.
Moberg GP. 1996. Suffering from stress: An approach for evaluating the
welfare of an animal. In: Sandoe P, Hurnik F, eds. Proceedings of Welfare of Domestic Animals Concepts Series in Method of Measurement.
Acta Agriculturae Scandinavica. Sect A, Animal Science(Suppl)
27:46-49.
Moberg GP. 1999. When does stress become distress? Lab Anim
28:422-426.
Moberg GP. 2000. Biological response to stress: Implications for animal
welfare. In: Moberg GP, Mench JA, eds. Biology of Animal Stress:
Implications for Animal Welfare. Wallingford, Oxon, UK: CAB Int
(Forthcoming).
Morton DB, Griffiths PHM. 1985. Guidelines on the recognition of pain and
discomfort in experimental animals and an hypothesis for assessment.
Vet Rec 116:431-436.
NRC [National Research Council]. 1992. Recognition and Alleviation of
Pain and Distress in Laboratory Animals. Washington DC: National
Academy Press.
Romanovsky AA, Kulchinsky VA, Akulich NV, Koulchitsky SV, Simons
CT, Sessler DI, Gourine VN. 1996. First and second phases of biphasic
fever: Two sequential states of the sickness syndrome? Am J Physiol
271:R244-R253.
Rushen J. 2000. Some issues in the interpretation of behavioural responses
to stress. In: Moberg GP, Mench JA, eds. Biology of Animal Stress:
Implications for Animal Welfare. Wallingford, Oxon, UK: CAB Int
(Forthcoming).
Sandkuhler J. 1996. The organization and function of endogenous antinociceptive systems. Prog Neurobiol 5 0:49-81.
Selye H. 1950. Stress. Montreal: Acta.
Selzer Z, Dubner R, Shir Y. 1990. A novel behavioral model of neuropathic
pain disorders produced in rats by partial sciatic nerve injury. Pain
43:205-218.
Short CE, Van Poznak A. 1992. Animal Pain. London: Churchill Livingstone.
Smith JA, Boyd KM. 1991. Lives in the Balance. The Ethics of Using
Animals in Biomedical Research. Report of the Working Party of the
Institute of Medical Ethics. Oxford: Oxford Univ Press.

Soma LR. 1987. Assessment of animal pain in experimental animals. Lab

Anim Sci 37:71-74.
Tabo E, Jinks SL, Eisele JH Jr, Carstens E. 1999. Behavioral manifestations
of neuropathic pain and mechanical allodynia, and changes in spinal
dorsal horn neurons, following L4-L6 dorsal root constriction. Pain
80:503-520.
Takaishi K, Eisele JH Jr, Carstens E. 1996. Behavioral and electrophysiological assessment of hyperalgesia and changes in dorsal horn responses
following partial sciatic nerve ligation in rats. Pain 66:297-306.
Tricklebank MD, Curzon G, eds. 1984. Stress-Induced Analgesia. Chicester:
Wiley.
UFAW [Universities Federation for Animal Welfare]. 1989. Guidelines for
the recognition and assessment of pain in animals. Prepared by a working party for the Association of Veterinary Teachers and Research Workers. Vet Rec 118:334-338.
Wallace J, Sanford J, Smith W, Spencer V. 1990. The assessment and

control of the severity of scientific procedures on laboratory animals.

Report of the Laboratory Animal Science Association Working Party.
Lab Anim 24:97-130.
Watkins LR, Mayer DJ. 1982. Organization of endogenous opiate and
nonopiate pain control systems. Science 216:1185-1192.
Watkins LR, Wiertelak EP, Goehler LE, Mooney-Heiberger K, Martinez J,
Furness L, Smith KP, Maier SF. 1994. Neurocircuitry of illness-induced
hyperalgesia. Brain Res 639:283-299.
Weiss JM. 1972. Influence of psychosocial variables on stress-induced pathology. In: Porter, R, Knight J, eds. Physiology, Emotion and Psychosomatic Dlness. Amsterdam: Elsevier. p 253-265.
Willis WD Jr. 1985. The Pain System. Basel: Karger.
Willis WD, Coggeshall RE. 1991. Sensory Mechanisms of the Spinal Cord.
New York: Plenum.
Willis WD, Westlund KN. 1997. Neuroanatomy of the pain system and of
the pathways that modulate pain. J Clin Neurophysiol 14:2-31.

Downloaded from http://ilarjournal.oxfordjournals.org/ at Universidad de Concepcion on June 6, 2016

Volume 4 1 , Number 2

2000

71