Evaluation of Four Predictive Algorithms for Intramammary Infection Status in Late Lactation Cows
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Evaluation of Four Predictive Algorithms for Intramammary
Infection Status in Late Lactation Cows
Sam Rowe,1 Amy Vasquez,2 Sandra Godden,3 Daryl Nydam4, Erin Royster3, Jennifer
Timmerman,3 Matthew Boyle,5
1The University of Sydney, Australia, 2DeLaval, USA, 3University of Minnesota, USA
4Cornell University, USA, 5Zoetis, USA
Sam Rowe BVSc MVM PhD DABVP
Senior Lecturer in Ruminant Medicine
University of Sydney
The University of Sydney Page 1
Acknowledgements and Funding The University of Sydney Page 2
Dry cow therapy The University of Sydney Page 3
Dry cow therapy strategies
– Blanket DCT INFECTED UNINFECTED
– 80% of farms in USA
The University of Sydney Page 4
Dry cow therapy strategies
– Blanket DCT INFECTED UNINFECTED
– 80% of farms in USA
– Treat all cows/quarters
The University of Sydney Page 5
Dry cow therapy strategies
– Blanket DCT INFECTED UNINFECTED
– 80% of farms in USA
– Treat all cows/quarters
– Selective DCT
– 10% of farms in USA
– Treat infected quarters only
– Benefits
• Reduced antibiotic use
• Lower drug costs
Algorithm-guided
The University of Sydney
SDCT Page 6
Recent clinical trial
• 1275 cows
• 7 herds from 4 sites
• Antibiotic use reduced by 55%
• No negative health impacts
Culture-guided
SDCT
Algorithm-guided
SDCT
The University of Sydney Page 7
Algorithms
Netherlands
Parity = 1: SCC < 150,000 cells/ml at the last test
Parity ≥ 2: SCC < 50,000 cells/ml at the last test
*Last test must be within 6 weeks of dry-off
United Kingdom
SCC < 200,000 cells/ml each of the last 3 tests
No clinical mastitis between the 3rd last test and dry-off
United States
SCC < 200,000 cells/ml at all tests
< 2 cases of clinical mastitis during whole lactation
New Zealand
Parity = 1: SCC < 120,000 cells/ml at all tests
Parity ≥ 2: SCC < 150,000 cells/ml at all tests
The University of Sydney Page 9
No clinical mastitis during whole lactation
Objectives
Evaluate four predictive algorithms for late lactation, cow-level
IMI, which was determined using standard bacteriology
Estimate the likely impact of each algorithm-guided SDCT
approach on dry cow antibiotic use in U.S. dairy herds
The University of Sydney Page 10
Original data collection
– Cross-sectional studies investigating risk factors for
intramammary infection
– Bedding
– Udder towels
J. Dairy Sci. 102:11384–11400 J. Dairy Sci. 102:11401–11413
https://doi.org/10.3168/jds.2019-17074 https://doi.org/10.3168/jds.2019-17075
© American Dairy Science Association®, 2019.
© American Dairy Science Association®, 2019.
Cross-sectional study of the relationships among bedding materials, bedding Cross-sectional study of the relationship between cloth udder towel
bacteria counts, and intramammary infection in late-lactation dairy cows management, towel bacteria counts, and intramammary
infection in late-lactation dairy cows
S. M. Rowe,1* S. M. Godden,1 E. Royster,1 J. Timmerman,1 B. A. Crooker,2 and M. Boyle3
1
2
Department of Veterinary Population Medicine, University of Minnesota, St. Paul 55108 S. M. Rowe,1* S. M. Godden,1 E. Royster,1 J. Timmerman,1 and M. Boyle2
1
Department of Animal Science, University of Minnesota, St. Paul 55108 Department of Veterinary Population Medicine, University of Minnesota, St. Paul 55108
3 2
Zoetis, Hager City, WI 54014 Zoetis, Hager City, WI 54014
ABSTRACT in used bedding and ALL-IMI varied by bedding type, ABSTRACT logistic regression. The quarter-level prevalence of IMI
with positive associations observed in quarters exposed was 19.6%, which was predominantly caused by non-
Objectives of this study were to (1) describe the in- to manure solids (OR = 2.29) and organic non-manure Because cloth udder towels (CUT) may function aureus Staphylococcus spp. (NAS; 10.2%) and SSLO
tramammary infection (IMI) prevalence and pathogen (OR = 1.51) and a negative association in quarters as a fomite for mastitis-causing pathogens, most ud- (5.1%). The predominant bacteria in CUT were Bacil-
profiles in quarters of cows approaching dry-off in US exposed to new sand (OR = 0.47). Findings from this der health laboratories offer towel culture services as lus spp. (median = 3.13 log10 cfu/cm2). Total bacteria
dairy herds, (2) compare IMI prevalence in quarters study suggest that quarter-level IMI prevalence in late- a tool to monitor towel hygiene. However, no studies count was not associated with odds of IMI (odds ratio
of cows exposed to different bedding material types, lactation cows is low in US dairy herds. Furthermore, have investigated if an association exists between bac- = 1.06), likely due to the predominance of Bacillus spp.
and (3) identify associations between bedding bacteria bedding material type may not be an important risk teria levels in CUT and udder health outcomes. The in CUT and low number of IMI caused by Bacillus spp.
count and IMI in cows approaching dry-off. Eighty factor for IMI in late lactation. Higher levels of bacteria objectives of this cross-sectional study were to (1) In contrast, counts of Staphylococcus spp. and SSLO
herds using 1 of 4 common bedding materials (manure in bedding may increase IMI prevalence at dry-off in describe associations between herd-level measures of were positively associated with odds of IMI caused by
solids, organic non-manure, new sand, and recycled general, but this relationship is likely to vary according towel bacteria count (ToBC) and quarter-level intra- NAS (odds ratio = 1.33) and SSLO (odds ratio = 1.45),
sand) were recruited in a multi-site cross-sectional to bedding material type. mammary infection (IMI) status in late-lactation cows, respectively. Of 12 CUT management practices evalu-
study. Each herd was visited twice for sampling. At Key words: intramammary infection, mastitis, dry (2) establish pathogen-specific target levels of bacteria ated, only the failure to use a dryer was identified as
each visit, aseptic quarter-milk samples were collected cow therapy, bedding, manure solids in CUT to aid the interpretation of towel culture re- a clear predictor of risk for a high ToBC (risk ratio of
from 20 cows approaching dry-off (>180 d pregnant). ports, and (3) identify laundering-related risk factors high coliform count = 8.17). Our study findings suggest
Samples of unused and used bedding were also col- for high ToBC. The study was conducted in 67 herds that CUT may act as a fomite for NAS and SSLO. We
INTRODUCTION
lected. Aerobic culture was used to determine the IMI from 10 dairy states in the United States that used recommend that herds aim to keep counts of Staphylo-
status of 10,448 quarters and to enumerate counts (log10 Cows acquire IMI during lactation, some of which can CUT. These 67 herds were originally recruited as part coccus spp. and SSLO in CUT below 32 cfu/cm2 (or 5
cfu/mL) of all bacteria, Staphylococcus spp., Strepto- persist through the dry period to affect udder health of a larger (80 herd) cross-sectional study of bedding cfu/in2), and that laundered towels be completely dried
coccus spp. and Streptococcus-like organisms (SSLO), in subsequent lactations (Green et al., 2002). To cure management. Each herd was visited once during De- in a hot air dryer.
The University
coliforms, Klebsiellaofspp.,
Sydney
noncoliform gram-negatives, these IMI, intramammary antimicrobial treatments are cember 2017 to April 2018 and quarter-milk samples
Page 11
Key words: towel bacteria count, cloth udder towel,
Bacillus spp., and Prototheca spp. in unused (n = 148) administered to cows at the time of dry-off (dry cow (n = 4,656) were collected from late-gestation (>180 intramammary infection, towel laundering, pre-milking
and used (n = 150) bedding. The association between therapy; DCT). However, there is interest within the d pregnant) cows (n = 1,313). Two recently laundered teat preparationMethods
– 80 herds originally recruited
from 10 states
– Selected for bedding type
– New sand (n=20)
– Reclaimed sand (n=21)
– Manure solids (n=20)
– Other organic (n=19) State Herds
CA 16
– Farms visited twice to enroll cows ID
IN
6
4
– Summer 2017 MI 5
– Winter 2017-18 MN 10
– 20 cows enrolled per visit NY 9
OR 1
TX 2
– Enrollment criteria WA 7
– Lactating WI 21
– Late gestation (> 180d pregnant)
The University of Sydney Page 12Analysis
CNA
MAC
MALDI-TOF MS
The University of Sydney Page 13Analysis
CNA
MAC
MALDI-TOF MS
The University of Sydney Page 14Analysis
MALDI-TOF MS
The University of Sydney Page 15Analysis The University of Sydney Page 16
Analysis
– Cow-level
– Test characteristics
– Kappa
– Sensitivity
– Specificity
1,594 cows from 56 farms
– Positive predictive value
– Negative predictive value
Agreement between IMI status and algorithm risk status
Sensitivity
Specificity
Cohen’s Kappa
The University of Sydney Page 17Algorithm Criteria for low risk (i.e. “test negative”) References
Parity = 1: SCC < 150,000 cells/ml at the last test
Netherlands Parity ≥ 2: SCC < 50,000 cells/ml at the last test Vanhoudt et al. (2018)
*Last test must be within 6 weeks of dry-off
Parity = 1: SCC < 120,000 cells/ml at all tests
New Zealand Parity ≥ 2: SCC < 150,000 cells/ml at all tests DairyNZ (2012)
No clinical mastitis during whole lactation
SCC < 200,000 cells/ml each of the last 3 tests Bradley et al. (2010),
United Kingdom
No clinical mastitis between the 3rd last test and dry-off Bradley et al. (2018)
SCC < 200,000 cells/ml at all tests
United States Rowe et al. (2020)
< 2 cases of clinical mastitis during whole lactation
Bradley, A., S. De Vliegher, M. Farre, L. Jimenez, T. Peters, E. de Leemput, and T. van Werven. 2018. Pan-European agreement on dry cow therapy. The Veterinary record
182(22):637.
Bradley, A., J. Breen, B. Payne, P. Williams, and M. Green. 2010. The use of a cephalonium containing dry cow therapy and an internal teat sealant, both alone and in
combination. J. Dairy. Sci. 93(4):1566-1577.
DairyNZ. 2012. Smart SAMM Technote 14. Vol. Accessed 2020.
Rowe, S., S. Godden, D. Nydam, P. Gorden, A. Lago, A. Vasquez, E. Royster, J. Timmerman, and M. Thomas. 2020. Randomized controlled non-inferiority trial investigating the
effect of 2 selective dry-cow therapy protocols on antibiotic use at dry-off and dry period intramammary infection dynamics. J. Dairy Sci. 103(7):6473-6492.
Vanhoudt, A., K. van Hees-Huijps, A. van Knegsel, O. Sampimon, J. Vernooij, M. Nielen, and T. van Werven. 2018. Effects of reduced intramammary antimicrobial use during
The
the dry University
period of Sydney
on udder health in Dutch dairy herds. J. Dairy. Sci. 101(4):3248-3260. Page 18Results The University of Sydney Page 19
Pathogens isolated at enrollment The University of Sydney Page 20
Pathogens isolated at enrollment
J. Dairy Sci. 102:11384–11400
https://doi.org/10.3168/jds.2019-17074
© American Dairy Science Association®, 2019.
Cross-sectional study of the relationships among bedding materials, bedding
bacteria counts, and intramammary infection in late-lactation dairy cows
S. M. Rowe,1* S. M. Godden,1 E. Royster,1 J. Timmerman,1 B. A. Crooker,2 and M. Boyle3
1
Department of Veterinary Population Medicine, University of Minnesota, St. Paul 55108
2
Department of Animal Science, University of Minnesota, St. Paul 55108
3
Zoetis, Hager City, WI 54014
ABSTRACT in used bedding and ALL-IMI varied by bedding type,
with positive associations observed in quarters exposed
Objectives of this study were to (1) describe the in- to manure solids (OR = 2.29) and organic non-manure
tramammary infection (IMI) prevalence and pathogen (OR = 1.51) and a negative association in quarters
profiles in quarters of cows approaching dry-off in US exposed to new sand (OR = 0.47). Findings from this
dairy herds, (2) compare IMI prevalence in quarters study suggest that quarter-level IMI prevalence in late-
of cows exposed to different bedding material types, lactation cows is low in US dairy herds. Furthermore,
and (3) identify associations between bedding bacteria bedding material type may not be an important risk
count and IMI in cows approaching dry-off. Eighty factor for IMI in late lactation. Higher levels of bacteria
The University of Sydney Page 21
herds using 1 of 4 common bedding materials (manure in bedding may increase IMI prevalence at dry-off inCows classified as ‘high risk’ by algorithms
50%
31%
63%
50%
The University of Sydney Page 23Test characteristics for all pathogens
0.13
0.12
0.12
0.05
Agreement with culture
(Cohen’s Kappa)
The University of Sydney Page 27Test characteristics for major pathogens
0.06
0.11
0.07
0.04
Agreement with culture
(Cohen’s Kappa)
The University of Sydney Page 28Test characteristics for major pathogens
So why do these algorithms
‘work’ in the field?
Bradley, A., J. Breen, B. Payne, P. Williams, and M. Green. 2010. The use of a cephalonium containing dry cow therapy and an internal teat
sealant, both alone and in combination. J. Dairy. Sci. 93(4):1566-1577.
Rowe, S., S. Godden, D. Nydam, P. Gorden, A. Lago, A. Vasquez, E. Royster, J. Timmerman, and M. Thomas. 2020a. Randomized controlled
non-inferiority trial investigating the effect of 2 selective dry-cow therapy protocols on antibiotic use at dry-off and dry period
intramammary infection dynamics. J. Dairy Sci. 103(7):6473-6492.
Rowe, S., S. Godden, D. Nydam, P. Gorden, A. Lago, A. Vasquez, E. Royster, J. Timmerman, and M. Thomas. 2020b. Randomized controlled
trial investigating the effect of 2 selective dry-cow therapy protocols on udder health and performance in the subsequent lactation. J. Dairy
Sci. 103(7):6493-6503.
Scherpenzeel, C. G., K. W. van den Heuvel-van den Broek, I. M. G. A. Santman-Berends, and G. van Schalk. 2020. Monitoring Udder Health
on Routinely Collected Census Data: Evaluating the Effects of Changing Antimicrobial Policy. Proc. 59th Annual meeting of the National
Mastitis Council:102-103.
Vasquez, A., D. Nydam, C. Foditsch, M. Wieland, R. Lynch, S. Eicker, and P. Virkler. 2018. Use of a culture-independent on-farm algorithm to
guide the use of selective dry-cow antibiotic therapy. J. Dairy. Sci. 101(6):5345-5361.
The University of Sydney Page 29Test sensitivity for selected pathogens
1.3% 1.4% 5.3% 1.1% 0.9%
The University of Sydney Page 30Conclusions
– Test performance for detection of
intramammary infection (considering all
pathogens) at the cow-level was poor
for all algorithms
– Test sensitivity and specificity were never concurrently high
– Kappas all in the ‘poor’ range
– This was also the case for major
pathogens
– Diagnostic sensitivity was better for
selected pathogens of interest including
Staphylococcus aureus and Streptococcus
uberis
The University of Sydney Page 31Conclusions
– There is no ‘perfect algorithm’
– Algorithms using more data points have high
sensitivity and lower specificity
– Eg. New Zealand and US algorithms
– A larger proportion of cows with IMI will receive treatment ✅
– A larger proportion of cows without IMI will receive treatment ❌
– Algorithms using less data points have low
sensitivity and high specificity
– Eg. Netherlands and UK algorithms
– A smaller proportion of cows with IMI will receive treatment ❌
– A smaller proportion of cows without IMI will receive treatment ✅
The University of Sydney Page 32Coalcliff, Australia The University of Sydney samuel.rowe@sydney.edu.au Page 33
You can also read
