title / A longitudinal study of faecal excretion of VTEC O157 in cattle to determine epidemiological patterns and risk factors associated with excretion
/ DEFRA
project code / OZ0138
Department for Environment, Food and Rural Affairs CSG 15
Research and Development
Final Project Report
(Not to be used for LINK projects)
Two hard copies of this form should be returned to:Research Policy and International Division, Final Reports Unit
DEFRA, Area 301
Cromwell House, Dean Stanley Street, London, SW1P 3JH.
An electronic version should be e-mailed to
Project title / A longitudinal study of faecal excretion of VTEC O157 in cattle to determine epidemiological patterns and risk factors associated with excretion
DEFRA project code / OZ0138
Contractor organisation and location / Veterinary Laboratories Agency, Woodham Lane, New Haw, Addlestone, Surrey
KT15 3NB
Total DEFRA project costs / £ 2,183,628.00
Project start date / 01/10/98 / Project end date / 31/01/05
Executive summary (maximum 2 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.
CSG 15 (Rev. 6/02) 3
Projecttitle / A longitudinal study of faecal excretion of VTEC O157 in cattle to determine epidemiological patterns and risk factors associated with excretion
/ DEFRA
project code / OZ0138
EXECUTIVE SUMMARY OF RESEARCH PROJECT OZ0138
A LONGITUDINAL STUDY OF FAECAL EXCRETION OF VTEC O157 IN CATTLE TO DETERMINE EPIDEMIOLOGICAL PATTERNS AND RISK FACTORS ASSOCIATED WITH EXCRETION
Objective
This 6-year project was aimed at providing epidemiological understanding of VTEC O157 in the cattle population of England and Wales. Sampling in cattle herds was done to determine the prevalence at farm-level and individual animal-level. A longitudinal study was then carried out to follow individual animals over time to investigate VTEC O157 excretion patterns and determine risk factors for VTEC O157 carriage. These results were used to develop and implement an intervention study with the intention of reducing the prevalence of VTEC O157 at farm-level.
Prevalence study
A total of 75 cattle herds (27 dairy, 33 suckler and 15 fattener) were sampled on one occasion between June and December 1999. A total of 38.7% (CI95: 28.1-50.4) of the farms were VTEC O157 positive with an individual animal prevalence on the positive farms of 10.3% (CI95: 5.6-14.9). No association between main farm type and prevalence of VTEC O157 was found. The highest prevalence was found in animals aged 2-12 months and the lowest prevalence in animals less than 2 months of age. Isolates revealed 11 phage types and vt2 was the most predominant toxin gene.
Longitudinal study
A minimum of 90 animals per farm on 9 VTEC O157 positive farms were sampled every month for 8-11 months. A maximum of four positive samples were collected from any one individual animal during the study, but 25.8% of animals were positive on more than one occasion. Some individuals excreted VTEC O157 on consecutive visits whilst others excreted intermittently. Several hygiene-related group-level risk factors were associated with excretion an individual animal in a multi-variate and multi-level model. These were high average muck score, presence of VTEC O157 in drinking water, access to milk and presence of an individual in the group which excreted over several months. Access to silage was found to reduce the risk of VTEC O157 excretion. Several phage types and several PFGE types were isolated from most farms and most animals. However, one phage type and PFGE profile seemed to predominate on each farm indicating persistence of the same strain.
Intervention study
A pilot intervention study was designed to investigate the effectiveness of three intervention packages described below. A total of 255 farms were screened for VTEC O157 of which 82 farms were positive. On the 255 farms, cattle which have access to springs at grazing, had less risk of VTEC O157 carriage. Assessing the bedding daily also seemed to reduce the risk. At a group-level a very dry bedding reduced the risk of the group being VTEC O157 positive and so did a lower number of animals in the pen. No management factors for groups kept on fields were associated with VTEC O157.
A total of 57 of the positive farms participated in the intervention study and were randomly allocated into three intervention groups and a control group. The study was conducted on an ‘intention to intervene’ basis. All intervention groups applied biosecurity measures at farm-level (closed herd policy, no contact with stock from neighbouring farms, a six week period between spreading of manure/dirty water/slurry on fields and grazing). Intervention A was aimed at keeping the bedding dry and avoiding transmission from other enclosures, intervention B kept a strict clean water and feed policy and intervention C applied both intervention A and B. The opportunity was also taken to assess the effect of these interventions on Campylobacter spp. and Eimeria spp.
The sample sizes were too small to show any significant differences in reduction of E. coli O157 between the intervention groups. However, keeping the bedding dry, using enclosure-specific boot dip and overcoat, and keeping other animals away from cattle enclosures (intervention A) appeared to reduce the prevalence of E. coli O157 quicker than the other intervention groups and the control group. Enclosure hygiene (intervention A) reduced the presence of Eimeria species significantly faster than the control group, but the interventions had no significant effect on the prevalence of Campylobacter spp.
Acknowledgements
We would like to thank DEFRA for funding this project and colleagues at CERA, VLA for managing database, doing analysis and report writing. Also a large thank you to VIOs and LTD staff in the VLA regional laboratories and all farmers, who have been involved in this project.
CSG 15 (Rev. 6/02) 3
Projecttitle / A longitudinal study of faecal excretion of VTEC O157 in cattle to determine epidemiological patterns and risk factors associated with excretion
/ DEFRA
project code / OZ0138
Scientific report (maximum 20 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.
CSG 15 (Rev. 6/02) 3
Projecttitle / A longitudinal study of faecal excretion of VTEC O157 in cattle to determine epidemiological patterns and risk factors associated with excretion
/ DEFRA
project code / OZ0138
SCIENTIFIC REPORT OF PROJECT OZ0138
List of contents
1. Prevalence study 3
2. Longitudinal study 5
3. Intervention study 7
4. Main findings 9
5. Key issues 10
6. Future work 10
7. List of figures and tables 3
8. References 11
List of figures and tables
Figure 1. Time line of OZ0138 3
Figure 2: Relationship between age and VTEC O157 excretion by cattle 4
Table 1: VTEC O157 phage types isolated from individuals [N=196] and by farm [N=29] 4
Table 2: Number of visits and samples collected by farm 6
Figure 3. Decrease in prevalence of E. coli O157 during the application of interventions 8
Figure 1. Time line of OZ0138
1. Prevalence study
01 Develop questionnaire and diary event forms for the prevalence survey and conduct pilot sampling of herds.
02 Recruit random sample of herds, brief herd owners and herdsmen on the requirements for the survey and formulate sampling time-table for each herd.
03 Obtain estimates of animal and herd prevalence of faecal excretion in cattle to examine the validity of the current working hypothesis that herd infection is relatively common, and the animal prevalence is low, but age-related.
The aim of this study was to determine the age-related prevalence of VTEC O157 in cattle in England and Wales.
Farms were randomly selected from the VETNET database. Selection was restricted to farms in England and Wales with a total herd size of 70 or more cattle as recorded on VETNET. This represented more than 80% of the cattle on holdings in England and Wales recorded in the June Agricultural Census 1997. Milk producer-retailers, cattle dealers and ‘open’ farms were excluded, as were farms with inadequate handling facilities. The remaining farms were then contacted by the closest regional VLA laboratory and assessed for eligibility. A total of 75 cattle herds (27 dairy, 33 suckler, 15 fattener herds) were recruited.
Enrolled farms were visited on a single occasion between June and December 1999.
In total, 4663 rectal faecal samples were collected from randomly selected cattle of all ages on the participating farms. A total of 219 VTEC O157 isolates were recovered, yielding an individual animal prevalence of VTEC O157 excretion of 4.2%, [CI95% 2.0 – 6.4] adjusted for cluster effect of groups, farms and region. In positive herds, the prevalence of individual animals excreting VTEC O157 was 10.3% [CI95% 5.6 – 14.9], but ranged from 1.2 - 51.4%. VTEC O157 were isolated from 38.7% [CI95% 28.1 – 50.4] of the herds tested.
Univariable analysis did not reveal any association between VTEC O157 excretion and herd type (p=0.49) or month of isolation (p=0.35).
VTEC O157 was found most frequently in animals aged 2 – 12 months, thereafter, a decline in the prevalence of infection as age increased was observed. Calves under 2 months of age were least likely to be infected (fig. 1).
Figure 2: Relationship between age and VTEC O157 excretion by cattle on all farms (A) and on positive farms only (B).
The VTEC O157 isolates belonged to 11 recognised phage types (PTs). A further 3 strains were present that reacted with the phages in the typing scheme, but did not conform to recognised types. The PTs 4, 34 and 2 were isolated from over a fifth of positive farms, whilst PTs 43, 8 and 34 accounted for the greatest number of isolates (Table 1).
Table 1: VTEC O157 phage types isolated from individuals [N=196] and by farm [N=29]
Phage type / Individual / FarmNo. / %1 / No. / %2
1 / 10 / 5.1 / 2 / 6.9
2 / 15 / 7.7 / 6 / 20.7
4 / 21 / 10.7 / 7 / 24.1
8 / 32 / 16.3 / 5 / 17.2
14 / 3 / 1.5 / 1 / 3.4
21/28 / 13 / 6.6 / 2 / 6.9
31 / 3 / 1.5 / 1 / 3.4
32 / 20 / 10.2 / 5 / 17.2
34 / 30 / 15.3 / 7 / 24.1
43 / 37 / 18.9 / 1 / 3.4
47 / 1 / 0.5 / 1 / 3.4
RDNC 1 / 6 / 3.1 / 1 / 3.4
RDNC 2 / 4 / 2.0 / 1 / 3.4
RDNC 3 / 1 / 0.5 / 1 / 3.4
Total / 196 / 29
1 Proportion of individuals
2 Proportion of herds
A single PT was recovered from 20 (69%) herds, whilst 2 PTs were isolated from 7 farms, and 3 and 4 different PTs were identified on one farm each. The predominant toxin gene was vt2 (99.5%), but both vt1 and vt2 genes were present in 25.0% (49) of the isolates.
03a. Validation of FOSS Enzyme Immunoassay (EIA) test to use for further screening.
The purpose of this study was to evaluate the performance of the FOSS EIA for VTEC O157 as a screening tool, when used to examine samples of bovine faeces for the presence of this organism. All samples from the prevalence study were analysed by FOSS EIA and the results compared to the ‘gold standard’ isolation by culture method used by VLA.
The true sensitivity of the FOSS EIA was 68.1% meaning that approx. 3 in 10 VTEC O157 will not be detected by this method. The specificity was 77.3% indicating that 78 out of 100 negative samples will be correctly identified as negative.
Predictive values depend on prevalence in the population and because of the difference in prevalence, a predictive value was calculated for all cattle and another one for cattle of 3-24 months of age. The positive predictive value for all cattle was 13.0%. For cattle at the age of 3-24 months, the positive predictive value was 26.5% , as the prevalence of excretion is higher in this age group. The negative predictive value was 95.6% for 3-24 months and 97.2% for all cattle. This means that of 97 out of 100 negative samples will be identified as negative by the FOSS EIA test. Agreement between the two tests has a kappa coefficient of 0.14, which implies that the agreement is poor. However, the two tests agree more often than would occur by chance (p<0.001).
The FOSS EIA was cheaper and faster than the culture method. Results could be obtained after two hours and it cost £5.70 per samples compared to £14.49 for the culture method. However, the sensitivity of this method was judged to be unacceptable to use for the following studies, since it would fail to identify 3 out of 10 VTEC O157 isolates.
2. Longitudinal study
04 Design and conduct a longitudinal study to investigate the validity of the risk factors, suggested in the scientific literature and those which emerge from the prevalence survey, for excretion of VTEC O157.
The aim of this study was to identify risk factors associated with persistence of excretion of VTEC O157 by individual cattle.
Twelve (36.4%) VTEC O157 positive farms from the prevalence study were enrolled in the longitudinal study. The first visit was conducted on the 27th of March 2000 and the last visit on 21st February 2001, yielding between 8-11 visits per farm. One farm was dropped after visit 6, because the farmer failed to provide the data needed for the study.
Monthly rectal samples were collected from up to 90 animals under 24 months of age. The animals were randomly selected at the first visit and sampled throughout the study.
In addition, three environmental samples were created by collecting faeces from floor faecal pats and pooling 5 samples into one during each visit. Surface water and biofilm from drinking troughs were also collected from enclosures of the sampled stock at each visit (table 2).
Table 2: Number of visits and samples collected by farm
Farm / Visits / Number of samples collectedNumber / Date of first visit / Animals / Individual / Environmental
in cohort / samples / Water / Biofilm / Pooled
1 / 8 / 23 May / 99 / 547 / 35 / 35 / 35
2 / 9 / 3 May / 278 / 809 / 53 / 53 / 53
3 / 10 / 11 May / 121 / 568 / 27 / 27 / 27
4 / 10 / 11 April / 59 / 325 / 34 / 17 / 35
5 / 9 / 8 May / 93 / 536 / 23 / 23 / 23
61 / 61 / 2 May / 174 / 567 / 30 / 30 / 30
7 / 10 / 11 April / 122 / 758 / 50 / 50 / 50
8 / 11 / 27 March / 79 / 700 / 38 / 37 / 37
9 / 10 / 17 April / 264 / 764 / 49 / 49 / 49
10 / 11 / 10 April / 136 / 1043 / 54 / 54 / 54
11 / 10 / 26 April / 54 / 364 / 32 / 32 / 32
12 / 8 / 27 June / 44 / 310 / 36 / 36 / 36
TOTAL / 112 / 1521 / 7291 / 1368
1 Farm dropped due to inadequate records (last visit on 2nd October 2000)