CHAPTER
5
THE PROBLEM OF SPECIES DIFFERENCES
Print
A major weakness of animal experiments conducted for medical
research is that differences exist between species which can
make results from one type of animal inapplicable to another.
Thus, pursuing a line of research on animals can produce conflicting
or confusing results, of unknown relevance to human beings.
This can have serious implications, at the worst misleading
researchers about the causes and characteristics of human illnesses
and delaying medical progress.
Mammals share many characteristics, but there are also critical
species differences at the levels of anatomy, physiology, pharmacology,
biochemistry and genetics, which may occur in any body system
or organ. Some of these variations are known, and perhaps can
be taken into account; but others, such as reactions to novel
drugs or the function of an area of the brain, are not yet discovered,
and in these cases the results from animal experiments can be
seriously misleading.
Species differences between cats, dogs, other animals and humans
are frequently reported in the scientific press, and this chapter
documents some important examples.
Species differences and experiments on cats
There are many obvious but nevertheless significant anatomical
and behavioural differences between cats and humans. Cats are
small quadrupeds, whilst humans are large bipeds. This is relevant
to the way the nervous systems of the two species have developed
to control limb movement. Humans are omnivorous; but cats are
true carnivores and have specialised senses of hearing and sight
for hunting. This is reflected in the structure and organisation
of their brains. Cats spend a large part of their time asleep,
and the remainder is normally spent patrolling, hunting and
marking territory.
Underlying these species variations, and beyond them, there
are many more subtle characteristics which distinguish cats
from humans and are significant in assessing the value of cat
experiments to human medical progress.
The brain and nervous system: sleep, pain modulation, vision,
olfaction, organ innervation and control of limb movement
In France and elsewhere, cats are widely used in neurological
research – all of the Case Studies using cats in this
Report are experiments on the brain and nervous system (see
Chapter 6). Yet there are many documented differences between
species in the organisation, structure and pharmacology of the
nervous system.
For example, the raphe nuclei in the brains of rats, cats and
humans – involved in numerous brain functions including
sleep and the control of pain – vary between the species
in the distribution of peptide-using brain cells and fibres
(43).
Neurons implicated in the regulation of sleep in the preoptic
area of the brain are clustered together in one region in mice
and rats. The location of this region differs slightly even
between rats and mice; but in cats, these sleep-active brain
cells are not located in a cluster but are diffusely scattered
in the preoptic area (44). Case Studies Twelve and Thirteen
using cats are examples of sleep research whose results may
not be relevant to humans because of species variations.
Another region of the brain, called the periaqueductal grey
matter, is involved in pain modulation and may play an important
role in the pain caused by migraine (see below) (45). However,
in this area of the brain, human neurons have more connecting
branches (dendrites) than the neurons of cats, which have fewer
dendritic branches but more dendritic spines (46). This suggests
a different complexity of local circuits between the two species
in this brain area.
Cats have been used in vision research for decades, despite
the major variations in the development and structure of the
human and feline visual systems (see
Case Study Eleven in Chapter 6). The night vision of cats
is particularly good due to a reflective tapetum lucidum in
the eye, not present in the human eye. The cat does not have
a macula or fovea, two regions of the retina of primary importance
for human vision (47). In his speech accepting the Nobel Prize
for vision research (48), Professor Torsten Wiesel stated that
the critical period in the development of the brain’s
visual function in young animals varies among species. In cats
the critical period lasts three to four months, whereas in human
children the duration may be five to ten years. Many studies
of visual deprivation have been conducted in cats, as well as
monkeys. In cats, visual deprivation can lead to loss of a specific
class of brain cells in the lateral geniculate nucleus of the
brain; but this was not found in visually deprived monkeys (49).
The cat’s well-developed sense of smell (olfaction) depends
on a separate, specialised pattern of arterioles supplying the
olfactory region of the brain. These arterioles are quite different
in origin and course from comparable blood vessels in the human
brain (50).
Maps of nerve inputs to the brainstem in rats, primates, and
cats, when compared, reveal distinct species differences in
organisation (51). Even at the microscopic level, important
differences have been observed in the ultrastructure of the
spinal cord in primates, compared with that in cats and rats
(52). Species differences occur in the density of innervation
of important organs and glands, such as the liver and pancreas.
The density of pancreatic nerve fibres in cats is much greater
than in dogs or humans (53). The general density of innervation
varies markedly in the parathyroid glands of chickens, rats,
guinea pigs, cats, dogs and sheep (54). Marked species variation
has also been reported in the density of innervation within
the cat, dog and human liver (55).
Limb co-ordination is directly relevant to the understanding
of neural control of movement. Inter-limb co-ordination patterns
are species-specific, and there are significant differences
between primates and cats (56). With regard to brain control
of upper limb movements, cats have no direct cortico-motoneuronal
connections in the spinal cord unlike primates (including humans)
. Differences have also been found in the physiology of nerves
controlling precision movements in cats and humans (58). Despite
these variations between cats and humans, cats are still used
in experiments to study nerve repair in the limb.
Haematology
Analyses of the blood of various animals have revealed significant
variations between species. Platelet aggregation sensitivity,
which appears to be related to atherosclerosis, varies widely
among different animals. A study of platelet aggregation found
sensitivity between species varied almost 20-fold, with humans
having the most sensitive platelets. Some of the most resistant
platelets were found in cats (59).
Wide variations have been found in the activity of an enzyme,
GABA-T (60), in blood platelets from six different species,
including cats, dogs and humans . Analysis of blood clotting
factors and fibrinolysis parameters in 11 species, including
cats and dogs, revealed marked dissimilarities between the human
clotting system and those of different animal species (61).
The change in plasma pH of oxygenated and deoxygenated blood
(the Haldane effect) is distinctly greater in cats and dogs
than in humans (62). A study of plasma from 11 different mammals,
including cats and dogs, indicated that plasma amine oxidase
(PAO), an enzyme assumed to exist in the serum of all mammals,
was not always present; and the substrate-specificity of the
enzyme varied within the species tested (63).
Disease ‘models’: stroke, Parkinson’s
disease, epilepsy, migraine and HIV
In medical research, cats are used as ‘models’ for
a variety of illnesses. Predominantly, these ‘models’
involve artificially inducing the symptoms of the human condition
whilst failing to replicate the underlying cause. The progression
of the disease can also be dissimilar to that of the human condition.
Animal ‘models’ can therefore seriously mislead.
For example, around the world, cats have been used in stroke
research. This involves surgically blocking a major brain artery,
causing brain damage similar to – but also different from
– a stroke in humans. The damage caused by experimental
artery blockage is known to vary according to the species of
animal used, and even between different breeds of the same species.
Moreover, the reaction of cats to an experimental drug to treat
stroke are unlikely to predict accurately the drug’s effect
on a human patient. During decades of research into stroke using
animals, numerous neuroprotective drugs appeared to work in
animals – but none has been safe or effective in humans
(64). A major reason for this failure is differences between
species.
Monkeys injected with the toxin MPTP develop a long-term movement
disorder which resembles (in some ways) human Parkinson’s
disease, and they are used for research into this condition.
But when cats are injected with MPTP they recover normal movement
control after two or three weeks and unlike monkeys, despite
further injections of MPTP, they do not develop a permanent
parkinsonian disorder (65).
Pancreatitis has been induced artificially in cats by passing
acid along the main pancreatic duct; and hydrocephalus has been
mimicked by injections of kaolin into the brain. Epileptic seizures
are easily induced in cats by the administration of penicillin,
but not in rats (66). Relying on information from such artificial
conditions which, moreover, often vary from species to species,
could seriously mislead researchers about the true causes and
etiology of these life-threatening disorders.
Migraine appears to be a specifically human problem, yet researchers
around the world (including, occasionally, in France) create
an artificial condition in cats. This involves anaesthetising
cats and electrically stimulating parts of their brain; they
are usually killed afterwards. Some successful drugs have been
developed using cat experiments; but the cat ‘model’
sheds no light at all on the cause of migraine in humans, so
that there has been very little research into ways of preventing
migraine attacks. Moreover, brain cells in the periaqueductal
grey matter, a region of the brain which may play an important
role in the pain caused by migraine, shows species differences
between cats and humans (67).
Even when cats do suffer from similar diseases to humans, there
are often differences in the cause, progression or clinical
symptoms. For example, cats infected with feline immunodeficiency
virus (FIV) develop cat AIDS. But FIV is structurally different
in many ways from the human virus, HIV, being more closely related
to similar viruses in cows and horses (68). Because of differences
in the viruses (FIV and HIV) and in the species (cats and humans),
the responses of cats to FIV and to experimental therapies are
unlikely to reliably predict the reaction of humans. Indeed,
it was a French study which demonstrated differences, between
HIV-infected humans and FIV-infected cats, in the control of
programmed cell death in white blood cells (lymphocytes) (69).
These blood cells play a highly significant role in the development
of AIDS.
Cats can suffer from the same genetic defect and lack of dystrophin
that causes muscular dystrophy in humans. Whilst humans with
this condition suffer muscle wasting, cats experience no disability
because their muscles regenerate (70). Diabetes mellitus in
cats and humans has similar symptoms, although it is very rare
in cats. Unlike humans, however, a genetic component in diabetes
has not yet been confirmed in cats, who have an unusual energy
metabolism (71).
Drug development and testing: anti-motion sickness and analgesia
Although animals are widely used in drug development and testing,
differences in physiology and metabolism mean that species can
vary enormously in the way they react to drugs, making them
unreliable models in the fields of pharmacology and toxicology.
For example, attempts to develop an animal model for anti-motion
sickness drugs, using cats and dogs, have been unsuccessful.
Scopolamine, probably the best single anti-motion sickness drug
for humans so far, has no effect in dogs, and findings in cats
are not definite (72).
Two of the most commonly used analgesics in humans, paracetamol
and aspirin, are toxic to cats (73). Differences in drug metabolism
have implications for the welfare of animals used in experimental
procedures. For example, it is difficult to provide safe analgesia
in cats. Non-steroidal anti-inflammatory drugs are poorly metabolised
in cats, so can only be used with caution. If used as analgesics
for cats, opioid pain-killers must be applied with care to prevent
over-dosing (74). In a number of Case Studies in this Report
(see Chapter 6), details of the pain-killers used for cats were
not provided in the scientists’ published papers. This
is poor practice, and a matter of some concern given the difficulties
of ensuring safe, effective analgesia in this species (see Chapter
1).
Species differences and experiments on dogs
Cardiovascular research
Dogs are widely used in cardiovascular research, not only by
academic scientists studying the structure and function of the
heart and blood vessels, but also by pharmaceutical companies
when testing drugs to check that there is no toxicity to the
cardiovascular system. Case studies One, Two, Four and Five
in this Report (Chapter 6) are cardiovascular experiments.
However, numerous reports have detailed important dissimilarities
between human and dog hearts, blood vessels and circulation.
A recent anatomical study comparing the atrio-ventricular junction
of human and dog hearts reported gross differences in anatomy
and structural differences in the conduction system (75). An
analysis of a crucial enzyme, cytochrome C oxidase, revealed
organ-specific forms of the enzyme in dogs depending on whether
the enzyme was from the heart or the liver. Organ-specific differences
of this kind were not seen in human tissue (76).
Researchers have detected variations between humans and dogs
in the effect of gravity on blood circulation in the lung according
to body position (e.g. supine versus prone positions). They
attribute the differences to the fact that dogs are quadrupeds
and humans walk upright (77). Yet dogs have often been used
as ‘models’ for humans in this kind of research.
There are also species differences in microcirculation to the
trachea and bronchi, which may be relevant to asthma research:
the subepithelial capillary network is dense in species such
as sheep and dog, but relatively scanty in rabbits and humans
(78).
There are considerable and significant species variations in
the activities of blood vessels (relevant to Case Study Two
in Chapter 6). For example, human basilar arteries from the
brain are relaxed by bradykinin, whereas dog basilar arteries
contract (79). Contractions induced by hypoxia (lack of oxygen)
in the human coronary artery in vitro are independent of the
endothelium (the lining of the vessels), and suppressed by the
drug indomethacin. However, dog coronary artery contractions
are weakened by removal of the endothelium, and are not influenced
by indomethacin (80).
Levels of the enzyme superoxide dismutase (SOD) in the walls
of coronary arteries vary widely among species. Human coronary
artery contains an average 6440 units of extracellular SOD per
gram, while in dogs the figure is only 160 units per gram (81).
This suggests that dogs and humans would differ greatly in their
susceptibility to damage caused by superoxide radicals.
There is significant evidence that high blood pressure seen
in obese humans is associated with insulin resistance and excessive
insulin in the blood (hyperinsulinemia). However, experiments
in dogs found no evidence that hyperinsulinemia raises blood
pressure. In fact, hyperinsulinemia in dogs actually lowered
their blood pressure, whereas in rats it raised blood pressure.
The authors admit that the implications for humans are unclear
(82).
Moreover, results obtained in dogs by examination of microvascular
blood supply of the spleen cannot reliably be transferred to
humans, since the morphology and function of the spleen differ
greatly between species (83). And controversial data concerning
the cardioprotective role of kinins, natural chemicals found
in the blood, may be due to a striking species variation in
the metabolism of kinins in serum in dogs, rats, rabbits and
humans (84).
Dogs are commonly used in tests of novel drugs to eliminate
those which are toxic to the cardiovascular system. One reason
that dogs are used is because they are especially sensitive
to cardiotoxicity, but this can also mean that drugs which were
not marketed because of this problem in dogs, might actually
have been safe in humans. A report which analysed the value
of cardiovascular tests in dogs for drug development suggested
that information relevant to humans, over and above that obtained
from other tests, was not provided by dog studies (85).
Drug metabolism and toxicity
Dogs are used frequently in drug studies, especially in tests
for absorption, distribution, metabolism and excretion (ADME)
and as the non-rodent species in repeat-dose toxicity tests.
Yet, there are widely documented important dissimilarities in
ADME and toxicity results between humans and dogs.
For example, the plasma half-life of a drug (a measure of how
long it stays in the bloodstream) is extremely important in
calculating safe doses and in interpreting the outcome of toxicity
tests. Table 5 below shows that there can be very large discrepancies
in this parameter – up to fifteen-fold – between
dogs and humans (86).
Table 5 Plasma half-lives of various drugs in humans and dogs
________________________________________________________
Plasma half-lives (hours) in humans and dogs
Drug |
Humans |
Dogs |
Antipyrine |
12 |
1.7 |
Digitoxin |
216 |
14 |
Digoxin |
44 |
27 |
Hexobarbitol
|
6 |
4.3 |
Meperidine |
5.5 |
0.9 |
Phenylbutazone |
72 |
6 |
Tromexan |
6 |
21 |
___________________________________________________
Inter-species variations in serum binding can profoundly affect
the therapeutic potential of certain drugs. Camptothecin was
a novel anti-cancer drug which was outstandingly successful
in animal trials, but has only a modest effect against human
cancers. Studies of serum from a number of species, including
dogs, found that the drug was uniquely bound in human serum
where it was converted into a biologically inactive form, making
it far less effective as an anti-cancer agent in humans (87).
A radiolabelled chemical which seemed effective in animal studies
for use in PET imaging encountered problems in human trials.
This was due to substantial species differences between dogs
and humans in the serum binding of the chemical (88).
Variations in drug ADME results are persistently encountered
in the development of new therapies. A study of the activity
of the important detoxification enzyme, glutathione-S-transferase,
in the red blood cells of eight different species including
dogs, found significant differences (89). A study of drug-metabolising
enzymes in dog, monkey and human intestines, undertaken by Merck
Research Laboratories, noted several species-specific differences
(90). For example, dog and human levels of three catalytic enzymes
differed significantly; and N-acetyltransferase was found in
human intestines but not in dog intestines.
The main anti-AIDS drug, called AZT, is metabolised much more
slowly in rat and dog liver cells than in those of monkeys and
humans (91). There are significant differences between rats,
dogs and monkeys in the clearance rates from the bloodstream,
and in the bioavailability, of a potential anti-AIDS drug called
indinavir (92). Furthermore, there are marked sex-related differences
in the clearance of indinavir in rats and dogs, but not in monkeys.
Contradictory results from different animal species make it
difficult to predict a drug’s disposition in humans (93)(see
Case Study Four in Chapter 6).
Because the processes of absorption, distribution, metabolism
and excretion vary extensively from species to species, studying
the fate of a drug in animals can fail to predict side effects.
For example, a comprehensive programme of animal studies in
mice, rats, hamsters, guinea-pigs, rabbits, dogs and rhesus
monkeys completed prior to human trials of the heart drug, amrinone,
failed to predict effects on the blood (thrombocytopenia) that
occurred in up to 20 per cent of patients receiving long-term
treatment (94).
The most common reason for patient withdrawal from treatment
with tamoxifen, an anti-cancer drug, is nausea and vomiting.
Yet even high doses given to dogs did not produce evidence of
vomiting (95). Mitoxantrone was synthesised in the hope of providing
an anti-tumour drug which would be less toxic to the heart.
Screening in beagle dogs did not demonstrate heart damage, but
the drug caused cardiotoxicity in patients (96).
Drugs with dangerous side effects continue to reach the market,
despite animal testing. Since the infamous thalidomide tragedy,
more than 120 drugs have been withdrawn for reasons of side
effects in humans in the UK, USA, France and Germany (97). Adverse
effects from drug treatments are a growing cause of illness
in patients.
In 2000, a review was published of data on 150 experimental
drugs, provided by twelve pharmaceutical companies (98). In
the review, the largest of its kind ever published, animal toxicity
test results were compared with toxicity to humans, as found
in clinical trials. The non-rodent tests, conducted primarily
in dogs, on average only predicted 63 per cent of the toxicities
later seen in clinical trials in humans. This is a surprisingly
low figure, and suggests that toxicity tests in dogs predict
the safety of medicines in humans relatively poorly.
Another report reviewed the published results of adverse effects
of novel drugs seen in dog tests (99). In 30 out of 60 studies,
significant toxicity was seen in dogs but also in rats. In the
remaining 30 dog studies, the adverse effects were only seen
in dogs (and not in rats). For 25 of these 30 studies, it was
concluded either that the effects in dogs would not occur in
humans (because of known species differences); or that the effects
could have been anticipated without dog tests (because of the
pharmacological action of the drug); or that the effects were
irrelevant to humans for other reasons, such as toxicity being
only seen at very high doses. The authors concluded that in
92 per cent of the cases they studied, use of dogs in drug toxicity
tests did not provide additional, relevant information.
Disease ‘models’: joint disorders, gastrointestinal
function and retinitis pigmentosa
Dogs have been used worldwide as ‘models’ of human
disorders of joints, including osteoarthritis and related cartilage
problems, as well as to study hip replacement surgery. To simulate
osteoarthritis, a gradually-developing condition usually found
in older humans, the cruciate ligament in the leg joint of dogs
is damaged, sometimes by a stab wound. This leads to joint instability,
and the cartilage changes which follow are considered to have
relevance to the human condition. However, in the last few years
it has been realised that animal models of osteoarthritis, including
in dogs, are unreliable for identifying new drug treatments
for humans (100). The value of these animal models has been
questioned even for developing new painkillers for osteoarthritis
(101).
There are also species differences in the biomechanical properties
of cartilage, including the meniscus cartilage of the knee,
which affect the load-bearing properties of the joints. Consequently
some scientists advise caution in extrapolating results from
animal ‘models’ to humans (102).
Dogs are often used as a ‘model’ to study bone adaptation
after a hip replacement operation. However, there are important
differences in the geometry of the femur (thigh bone) in dogs
compared to humans. This may explain why results from dog experiments
indicate greater bone loss than is actually found in patients
who undergo hip replacement (103). Dogs are also widely used
to determine the efficacy and safety of materials and designs
used in hip replacement operations. Results cannot be extrapolated
with confidence from dogs to humans because of differences between
the species in bone shape and angles, which complicate understanding
of stress transfer across the hip (104 - 105).
Research into human gastrointestinal function sometimes uses
dogs (e.g. Case Study Three in Chapter 6). But again, there
are species variations which can complicate the interpretation
of animal experiments. For example, the localization and the
density of receptors to peptide hormones in the upper gastrointestinal
tract vary between humans, dogs and rats (106). A comparison
of several species, including humans and dogs, found differences
in the gastrointestinal distribution of IAPP, a natural protein
found in the gut and pancreas (107). IAPP affects metabolism
and gut muscle contractions.
Retinitis pigmentosa is an inherited human condition which eventually
leads to blindness. A similar condition occurs naturally in
dogs and mice, but even so, there can be discrepancies in results
of experiments on different species. For example, mice with
the same genetic mutation as human patients, when treated with
a drug called diltiazem, showed reduced retinal degeneration.
The scientists therefore predicted that this drug might be useful
for human patients (108). However when diltiazem was tried in
dogs with retinitis pigmentosa, it had no beneficial effect
at all (109).
A better way of doing research
Species differences range from the glaringly obvious to the
most subtle biochemical variations, across all body systems,
but always with the same implication: they make it very difficult
to reliably apply findings in one species to another.
This is also true of biomedical research conducted on cats and
dogs. The value of such experiments is often exaggerated by
the scientific community, although a close study of the literature
tells a different story.
In evolutionary terms, dogs and cats are not close relatives
of humans. This means that they are dissimilar to us in many
ways, from the structure and function of their brains, to their
responses to drugs and other chemicals.
The problem of species variability in research can be overcome
by the application, instead, of humane methods selected for
their relevance to humans, such as clinical and human volunteer
studies, epidemiological research, human cell and tissue culture,
sub-cellular and molecular research and post-mortem studies
of cadavers. Such approaches will benefit people as well as
animals.
(43) R Covenas et al [2001] Reviews in Neurology, 33, 131-137
(44) D Wagner et al [2000] Sleep Research Online, 3, 35-42
(45) YE Knight & PJ Goadsby [2001] Neuroscience, 106, 793-800
(46) M Gioia et al [1998] Anatomical Record, 251, 316-325
(47) NC Buyukmihci [1990] Proceedings of the First International
Medical Conference: Future Medical Research Without the Use
of Animals: Facing the Challenge, 1990, p57. Publ. CHAI
(48) Anon. [1982] Nature, 299, 583-591
(49) JB Levitt et al [2001] Journal of Neurophysiology, 85,
2111-2129
(50) E Sztamska & B Goetzen [1997] Folia Neuropathologica,
35, 60-68
(51) HH Molinari et al [1996] Journal of Comparative Neurolology,
375, 467-480
(52) SM Carlton et al [1996] Journal of Comparative Neurology,
371, 589-602
(53) C Sternini et al [1992] Cell & Tissue Research, 269,
447-458
(54) L Luts & F Sundler [1994] Regulatory Peptides, 50,
147-158
(55) YS Lin et al [1995] Comparative Biochemistry & Physiology
A Physiology, 110, 289-98
(56) LJ Shapiro et al [1997] American Journal of Physical Anthropology,
102, 177-186
(57) K Nakajima et al [2000] Journal of Neurophysiology, 84,
698-709
(58) N Kakuda et al [1996] Journal of Physiology, 492, 921-929
(59) KC Hayes & A Pronczuk [1996] Comparative Biochemistry
& Physiology B Biochemical & Molecular Biology, 113,
349-353
(60) FM Sherif et al [1993] Comparative Biochemistry & Physiology
C, 104, 345-349
(61) HE Karges et al [1994] Arzneimittelforschung, 44, 793-797
(62) H Kiwull-Schone et al [1992] Advances in Experimental Medical
Biology, 316, 11-20
(63) S Ebong & WR Farkas [1993] Comparative Biochemistry
& Physiology C, 106, 483-487
(64) J De Keyser et al [1999] Trends in Neurosciences, 22, 535-540
(65) DS Rothblat & JS Schneider [1995] Neurodegeneration,
4, 87-92
(66) D Hategan et al [1995] Romanian Journal of Neurology &
Psychiatry, 33, 103-108
(67) M Gioia et al [1998] Anatomical Record, 251, 316-325
(68) RA Olmsted et al [1989] Proceedings of the National Academy
of Sciences USA, 86, 8088-8092
(69) AL Guiot et al [1997] International Journal of Immunopharmacology,
19, 167-179
(70) T Partridge [1991] Neuropathology & Applied Neurobiology,
17, 353-63
(71) K Froslund [1997] Alternatives To Laboratory Animals, 25,
183-185
(72) BS Cheung et al [1992] Journal of Clinical Pharmacology,
32, 163-175
(73) S Wolfensohn & M Lloyd [1995] Handbook of Laboratory
Animal Management and Welfare, p228. Publ. Oxford University
Press
(74) S Wolfensohn & M Lloyd [1995] op cit
(75) SY Ho et al [1995] Journal of Cardiovascular Electrophysiology,
6, 26-39
(76) D Linder et al [1995] Comparative Biochemistry & Physiology
B Biochemical & Molecular Biology, 112, 461-469
(77) RD Ross et al [1997] Clinical Science, 92, 81-85
(78) J Widdicombe [1996] Microcirculation, 3, 129-141
(79) K Schror & R Verheggen [1988] Trends in Pharmacological
Sciences, 9, 71-74
(80) N Toda et al [1992] American Journal of Physiology, 262,
678-683
(81) P Stralin et al [1995] Arteriosclerosis, Thrombosis &
Vascular Biology, 15, 2032-2036
(82) MW Brands & JE Hall [1992] Journal of the American
Society of Nephrology, 3, 1064-1077
(83) P Groscurth [1985] The Lancet, 2, 562
(84) A Decarie et al [1996] American Journal of Physiology,
271, 1340-1347
(85) CL Broadhead, M Jennings & RD Combes [1999] A critical
evaluation of the use of dogs in the regulatory toxicity testing
of pharmaceuticals. Publ. FRAME & RSPCA, England
(86) R Levine [1978] Pharmacology: Drug Actions and Reactions.
Publ. Little, Brown
(87) Z Mi & TG Burke [1994] Biochemistry, 33, 12540-12545
(88) CJ Mathias et al [1995] Journal of Nuclear Medicine, 38,
1451-1455
(89) JK Vodela & RR Dalvi [1997] Veterinary & Human
Toxicology, 30, 9-11
(90) T Prueksaritanont et al [1996] Drug Metabolism & Disposition,
24, 634-642
(91) F Nicolas et al [1995] Drug Metabolism & Disposition,
23, 308-313
(92) JH Lin et al [1996] Drug Metabolism & Disposition,
24, 1111-1120
(93) JH Lin et al [1996] Drug Metabolism & Disposition,
24, 1298-1306
(94) CT Eason et al [1987] Human Toxicology, 6, 436
(95) MJ Tucker et al [1984] In: Safety Testing of New Drugs
-- Laboratory Predictions and Clinical Performance, p 157-8.
Ed. Lawrence, McLean & Wetherall. Publ. Academic Press
(96) R Stuart-Harris et al [1984] The Lancet, 28 July, 219-220
(97) C Spriet-Pourra & M Auriche [1994] SCRIP Report: Drug
Withdrawal from Sale.
(98) H Olson et al [2000] Regulatory Toxicology & Pharmacology,
32, 56-67
(99) CL Broadhead, M Jennings & RD Combes [1999] A critical
evaluation of the use of dogs in the regulatory toxicity testing
of pharmaceuticals. Publ. FRAME & RSPCA, England
(100) WB van den Berg [2001] Current Opinion in Rheumatology,
13, 452-456
(101) KD Brandt [2002] Biorheology, 39, 221-235
(102) MD Joshi et al [1995] Journal of Biomedical Materials
Research, 29, 823-828
(103) DR Sumner et al [1990] Journal of Orthopedic Research,
8, 671-677
(104) RD Bloebaum et al [1993] Journal of Biomedical Materials
Research, 27, 1149-1159
(105) TY Kuo et al [1998] Journal of Biomedical Materials Research,
40, 475-489
(106) JC Reubi et al [1997] Gastroenterology, 112, 1197-1205
(107) H Mulder et al [1997] Peptides, 18, 771-783
(108) M Frasson et al [1999] Nature Medicine, 5, 1183-1187
(109) SE Pearce-Kelling et al [2001] Molecular Vision, 7, 42-47
-
Chiens & Chats
Cosmétiques
Méthodes
substitutives
-
