Arch Virol (1998) 143: 1155–1162
Humane endpoints are an objective measure of morbidity
in Venezuelan encephalomyelitis virus infection of mice
A. J. Wright and R. J. Phillpotts
D.E.R.A., Microbiology Department, Chemical and Biological Defence Sector,
Proton Down, Wiltshire, U.K.
Accepted January 8, 1998
Summary.Theclinical signsare describedof Venezuelanencephalomyelitis virus
(VEEV) infection in mice after both airborne and subcutaneous (s.c.) challenge.
Group clinical scores reﬂected the known pathogenesis of infection by both s.c.
and airborne challenge, and with epizootic and enzootic strains of VEEV. This
observation conﬁrms the speciﬁc relationship of the observed clinical signs to
VEEV infection. Within an experiment, those who are assessing the animals for
clinical signs must have a common understanding of their appearance, including
severity, and should be unaware of the allocation of treatments. If these conditions
are met, the progress of clinical signs may be used to determine objectively the
time of culling for humane endpoints.
Under the UK Animals (scientiﬁc procedures) Act 1986, there is a requirement to
keepthesuffering ofexperimentalanimalstoaminimum. Thismay beachievedby
reducing numbers, and considering means of replacement . However experi-
ments in the whole animal remain essential for assessments of drug and vaccine
efﬁcacy in which complex systems, dependent upon tissue architecture interact
dynamically. These processes cannot yet be replicated in vitro. The death of an
animal from infection, as an endpoint, is unnecessary if the purpose of an ex-
periment is to determine whether a drug or vaccine has some beneﬁcial effect.
Current legislation requires that animals should be culled, where possible, to
minimise suffering, i.e. that “humane endpoints” are used. In practice therefore
animals must be culled humanely when clinical signs reach a given, objectively
discernible level of severity. Furthermore information may be lost if death is the
only clinical sign considered. Statistical analyses may be more informative if
performed on raw data consisting of continuous or truncated variables, thereby
enabling a further reduction in the number of animals used.