Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
Test characteristics of differential cell counts to identify subclinical intramammary infections estimated by Bayesian latent class analysis
Van Werven, T.; Sietsma, S.;...
Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
Read
Objectives: Prudent use of antibiotics requires that the treatment of subclinical mastitis is based on the treatment of cows with a confirmed intramammary infection (IMI). Because subclinical mastitis is defined as an infection in absence of visible changes in milk or udder appearance, the use of a diagnostic test is needed. The three most commonly used tests are: 1) somatic cell count (SCC), 2) bacteriological culture (BC) and 3) polymerase chain reaction (PCR). All these tests are imperfect, and lack either sensitivity (BC) or specificity (PCR) or both (SCC). In this study, we investigate the performance of VETSCAN DC-QTM, an automated differential leukocyte device, which uses the different leukocyte counts in combination with the total SCC to diagnose IMI. A model that accounts for classification error is the latent class model, which allows estimation of the sensitivity (Se) and specificity (Sp) of different tests assuming no gold standard test.
The objective of our study is to determine the test characteristics of the VETSCAN DC-QTM to detect quarter level IMI in combination with quarter level SCC (QSCC) at a threshold of 100.000 cells/mL, BC and PCR in cows with a new elevated cow level SCC just after calving (≤ 30 days), early lactation (> 30 days) and in cows with an elevated SCC in the last 4 weeks before drying off.
Material & Methods: The VETSCAN DC-QTM has three different settings based on DIM of the animals to be tested. To validate the three different settings of the VETSCAN DC-QTM a total of 449 cows, i.e. 1796 milk samples on quarter level, were collected on 8 Dutch dairy herds. Inclusion criteria were a 4-weekly DHI- participation and high new IMI rate. Cows were selected on first elevated cow level SCC (≥ 200.000 cells/ mL) in current lactation or with elevated SCC within 4 weeks before dry off. Quarter samples were taken aseptically be- fore cluster attachment, within one week after DHI sampling. Within 6 hours, these samples were split and QSCC (Delaval DCC) and VETSCAN DC-QTM test were performed, BC and PCR samples were stored at -18°C and these tests were later performed at two different laboratories. Bayesian latent class models were built to determine Se and Sp of the 4 tests, assuming no gold standard test.
Results: Results of BC and PCR were classified as minor or major pathogens. Of the 1794 samples analysed 23% test- ed positive by BC, 42% tested positive by QSCC at a threshold of 100.000 cells/mL (QSCC>100), 8% was positive for a major pathogen in PCR, 68% was positive for a major and/or a mi- nor pathogen in PCR, and 41% tested positive by VETSCAN DC-QTM. The most commonly isolated pathogens by BC wereStreptococcus uberis (8% of all samples) and Coagulase neg- ative Staphylococcus spp. (6%).
Results from the latent class analysis model that estimated Se and Sp of identifying IMI caused by major or minor pathogens showed VETSCAN DC-QTMto have a Se varying from 0.64-0.90 and a Sp of 0.63-0.86 for the three different lactation periods. These estimates were almost similar to the test char- acteristics of standard QSCC>100. Overall test characteristics were substantially better than those from BC, which had very low Se (0.37) and PCR, which lacked Sp (0.51). For diagnosing IMI caused only by major pathogens, the Se of the VET- SCAN DC-QTM varied between 0.90-0.98, and Sp for “Early Lactation” and “30+ DIM” was 0.71 and 0.76, respectively, but Sp dropped to 0.27 in the category “Dry Off”.
Conclusion: Altogether, our results suggest that the VET- SCAN DC-QTM has good test characteristics for diagnosing intramammary infections, but the added value over an undifferentiated QSCC measurement remains to be demonstrated.
Usability of the VETSCAN DC-QTM will depend on the goal of the test. For screening cows with new IMI in early lactation in order to treat infections before they become clinical or chronic, it is a useful and convenient tool, with fair to near excellent Se and Sp at a quarter-level. For selecting cows at the end of lactation for selective dry cow therapy (SDCT) based on the presence of only major pathogens, the results of the VETSCAN DC-QTM might lead to a higher use of antibiotics than necessary, because of the low Sp.
Keywords: Differential cell count; dairy; subclinical masti- tis; Bayesian latent class analysis.
Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
About
Affiliation of the authors at the time of publication
Utrecht University, Faculty of Veterinary Medicine, Utrecht, Netherlands;
University Farm Animal Practice, Harmelen, Netherlands.
Comments (0)
Ask the author
0 comments