A total of 355 eligible cases from 542 questionnaires were identified which were not from an outbrea

1
ZOONOTIC TRANSMISSION OF CRYPTOSPORIDIUM - A
2-YEAR EPIDEMIOLOGICAL STUDY
Smith, R.P.1, Chalmers, R.M.3, Elwin, K.3,
Hadfield, S.3, Mueller-Doblies, D.2,
Clifton-Hadley, F.A.2, Watkins, J.4 and Giles,
M.2
(1) Centre for Epidemiology and Risk Analysis
(CERA), Veterinary Laboratories Agency Weybridge
(VLA), New Haw, Addlestone, Surrey, KT15 3NB,
United Kingdom (2) Food and Environmental Safety
(FES), Veterinary Laboratories Agency Weybridge
(VLA), New Haw, Addlestone, Surrey, KT15 3NB,
United Kingdom (3) UK Cryptosporidium Reference
Unit, NPHS Microbiology Swansea, Singleton
Hospital, Swansea, SA2 8QA, United Kingdom (4)
CREH Analytical, Leeds, LS18 4RS.

Aim This study investigated the relationship
between Cryptosporidium isolates of human origin
and isolates from animal and environmental
contacts in the two weeks prior to the onset of
symptoms. By analysing the Cryptosporidium
species and subtypes present in the human cases
and their animal contacts we addressed the
question of whether the human infection could
have been caused by zoonotic transmission.
Introduction Cryptosporidium is an important
protozoan parasite which can cause diarrhoeal
disease in humans. In healthy individuals it is
usually self-limiting, but can be fatal in
immunocompromised patients. There are two
principal species which cause human disease in
the UKCryptosporidium parvum and Cryptosporidium
hominis, which between them account for 96 per
cent of cases in approximately equal proportions.
C. parvum is considered to be a zoonotically
acquired infection, while C. hominis is generally
considered to be primarily spread within human
populations. Both species can be transmitted
directly or indirectly e.g. waterborne. The
relative risk of human infection through direct
or indirect contact with animals or animal faeces
during occupational or recreational activities is
largely unknown.
A total of 355 eligible cases (from 542
questionnaires) were identified which were not
from an outbreak, not associated with foreign
travel and had no other household members with
diarrhoea. Permission was sought to follow up
farm animal or environmental exposures. The
contact animal and environmental samples were
analysed for the presence of Cryptosporidium
oocysts by immunomagnetic separation and
immunofluorescence microscopy (IFM), PCR-RFLP at
the 18s rRNA locus, PCR at the dhfr locus and
analysis of part of the GP60 gene. Sequencing was
used to confirm speciation results and for GP60
subtyping.
Method
Between Nov 2004 and Nov 2006, 28 diagnostic
laboratories, in 41 local authorities within the
study areas of South West and East England and
Wales, submitted 1030 human Cryptosporidium-positi
ve faecal samples to the UK Cryptosporidium
Reference Unit for species/subtype identification
by PCR-RFLP and sequencing. Human cases were
asked by the Local Authority Environmental Health
Department to complete questionnaires, to
identify cases reporting farmed animal contact or
key environmental exposures (manure, slurry,
private water supply).
Results
.
Figure 1 Cryptosporidium species from eligible
human cases
Cryptosporidium species in subset of eligible
human cases
The table below shows the speciation results of
all human cases and those that were eligible for
the study, with or without risk exposure. Figure
1 shows the percentage breakdown of species for
eligible cases. Human cases reporting either farm
animal or environmental contact were
significantly more likely to have C. parvum than
the rest of the study population (Chi-squared
test Plt0.001).
Farm animal sampling results
Sampling consent was given by the farmers in
22/38 cases. 371 samples were collected from the
22 contact farms
88/355 (24.8) eligible human cases reported
farm/ environment contact and 38 gave permission
to contact the farmers for further sampling.
144/371 (38.8) farm animal samples were positive
for Cryptosporidium by IMS-IFM and/or PCR
analysis. Of those tested and positive by PCR,
47/63 (74.6) were C. parvum, 4 were C. bovis, 1
C. andersoni, 1 unusual type and 10 unclear
patterns.
Cryptosporidium species comparison
Of the 22 human cases which were followed up by a
farm visit, 18 were positive for C. parvum
(predominantly GP60 allele IIa)
The 5 matched GP60 subtypes were all IIaA17G1R1,
originating from 1 farm dog, 10 cattle and 2
sheep.
In 5 of the human cases, the human subtype
matched the animal subtypes
Conclusion Our data provides molecular evidence
of matched species and subtypes in humans and
animals, who had recent contact, to support the
evidence of a zoonotic transmission pathway of C.
parvum through direct or indirect contact with
farm animals
Acknowledgements The authors would like to thank
collaborating colleagues at VLA and CRU. This
study was funded by Defra under project OZ0407
For further information contact Centre for
Epidemiology and Risk Analysis An Executive
Agency of the m.giles_at_vla.defra.gsi.gov.uk Veterin
ary Laboratories Agency-Weybridge, Department
for Environment, Food r.smith_at_vla.defra.gsi.gov.
uk Addlestone, Surrey KT15 3NB, UK. Rural
Affairs Tel. [phone] www.defra.gov.uk/corpo
rate/vla