Proceedings of the 10th International Colloquium on Paratuberculosis

#244 A survey to estimate <strong>the</strong> herd-level prevalence <strong>of</strong> paratuberculosis in <strong>the</strong> dairy herd <strong>of</strong>
<strong>the</strong> United Kingdom using serum antibody
George Laurie Caldow, Fiona Burns, Elizabeth Mackay, Robin Sayers, Alasdair James Charles Cook
Scottish Agricultural College, United Kingdom; VLA, United Kingdom
A survey was conducted to estimate <strong>the</strong> prevalence <strong>of</strong> dairy herds in <strong>the</strong> United Kingdom that were infected
with paratuberculosis. This report details <strong>the</strong> serological study carried out in support <strong>of</strong> this objective. Herds
were selected at random from throughout <strong>the</strong> UK and 353 farmers were invited to participate. 136 (39%) <strong>of</strong> <strong>the</strong>
contacted farmers were recruited. Farms with a previous history <strong>of</strong> clinical paratuberculosis were more likely to
join <strong>the</strong> study. The herd size ranged from 57 to 413 milking cows. In each herd all animals <strong>of</strong> three years <strong>of</strong> age
and above were sampled (mean sample size <strong>of</strong> 101). In total blood samples from 13,691 cows were tested using
a commercially available enzyme-linked immunosorbent-assay (ELISA) to test for <strong>the</strong> presence <strong>of</strong> antibody
to Mycobacterium avium subspecies paratuberculosis (Map). The results <strong>of</strong> <strong>the</strong> survey will be reported and
related to <strong>the</strong> o<strong>the</strong>r parts <strong>of</strong> this study that involved faecal culture and examination <strong>of</strong> <strong>the</strong> bulk tank milk for <strong>the</strong>
presence <strong>of</strong> Map.
#245 The validity <strong>of</strong> repeated serological and culture results to determine <strong>the</strong> true infection
status for Mycobacterium avium subsp. paratuberculosis
Gerdien van Schaik, Wim Swart, Kees van Maanen, Maarten F. Weber
GD Animal Health Service, The Ne<strong>the</strong>rlands
There is no real gold standard for detecting <strong>the</strong> infection status for Mycobacterium avium subsp. paratuberculosis
(MAP) in live cattle. The likelihood <strong>of</strong> a (non-)infected status may be increased by repeated test results with
different diagnostic tests. The objective <strong>of</strong> our study was to determine <strong>the</strong> added value <strong>of</strong> repeated test results
to determine <strong>the</strong> real infection status for MAP.
Seventeen MAP infected herds were selected for <strong>the</strong> study. Twice a year all dairy cows in a herd were
tested in parallel with a serum-ELISA (Pourquier) and fecal culture on solid LJ medium . Eight rounds <strong>of</strong>
testing were carried out. The farmers only received <strong>the</strong> results for <strong>the</strong> first round <strong>of</strong> testing, <strong>the</strong> o<strong>the</strong>rs were not
disclosed. Hierarchical models were made in WinBUGS 14.0, in which <strong>the</strong> ELISA result was regressed against
<strong>the</strong> fecal culture result, <strong>the</strong> age <strong>of</strong> <strong>the</strong> cow, <strong>the</strong> round <strong>of</strong> testing and a random cow and herd effect.
The multilevel model showed that only <strong>the</strong> random cow effect (3.87(SE 0.36)) was significant and that
<strong>the</strong>re was no significant random herd effect (0.016 (SE 0.0460)). Thus, repeated observations on a cow were
correlated but cows within a herd were not correlated. The sensitivity <strong>of</strong> <strong>the</strong> ELISA in faecal culture negative
cows is very low (1-2%). The sensitivity increased with age and with increasing amounts <strong>of</strong> bacterial shedding
up to 89% for a heavy shedder older than 4 years. The specificity for non-shedding cows in infected herds
slightly decreased from 99.2% in heifers to 97.9% in cows older than 4 years.
The correlations for <strong>the</strong> currently ELISA positives (S/P>90) with <strong>the</strong> previous results were fairly low (0.4-
0.6) and only significant until two samples back (t-2). Therefore, infected herds should be tested annually.
193