Comparison of fecal pooling strategies for detection of Mycobacterium avium ssp. paratuberculosis in cattle.

In herds with typical moderate to low within-herd prevalence, testing for Mycobacterium avium ssp. paratuberculosis (MAP), the infectious agent of Johne's disease, will be more cost-effective if individual fecal samples are cultured in composite pools. However, sensitivity to classify a pool containing 1 or more positive individual samples as positive may depend on pool size and number of individual positive samples within a pool. Fecal samples collected from 994 dairy cows sampled at slaughter were cultured to detect MAP. Culturing was done both individually and as composite pooled samples using the TREK ESP Culture System II broth medium (Thermo Fisher Scientific, Trek Diagnostic Systems Inc., Cleveland, OH). Composite samples consisted of pools containing feces from 3, 5, 8, 10, or 15 cows. The number of individual fecal culture-positive cows within each pool ranged from 0 to 4. Culture of individual fecal samples detected MAP in 36 (3.6%) of the 994 cows. Individual samples that were detected within the first 50 d by TREK ESP Culture System II were more likely to lead to a positive pool result. In total, 840 pooled fecal samples were examined for presence of MAP, and of those, 272 pools actually contained feces from fecal culture-positive cows. The crude sensitivity (proportion of pools that contained at least 1 fecal-positive cow that tested positive) for pools of 3, 5, 8, 10, and 15 was 47, 67, 44, 59, and 39%, respectively. Across pools, an increase of the number of fecal culture-positive samples from 1 to 2 enhanced overall crude sensitivity from 44 to 71%. However, sensitivity did not further increase for pools with 3 or 4 fecal culture-positive samples (63 and 60%, respectively). Additionally, a simulation analysis assessing probability of pooled fecal samples being positive in herds of 50 and 100 cows was conducted. The simulation assumed that 1, 2, or 5 cows per herd were MAP fecal culture-positive and that pools of 5 and 10 were used. This low-prevalence herd simulation indicated that weighted mean herd probabilities of detecting a positive herd ranged between 52 and 99.3%, with the lowest probability for pools of 10 with 1 positive cow in the herd and the highest probability for pools of 5 with 5 positive cows in the herd. However, overall, pools of 5 and 10 had similar diagnostic capabilities, enabling cost savings by utilizing pools of 10.

Evaluation of milk ELISA for detection of Mycobacterium avium subspecies paratuberculosis in dairy herds and association with within-herd prevalence.

Cow-level milk ELISA results can be used to determine herd Mycobacterium avium ssp. paratuberculosis (MAP) status. Milk sample collection is minimally invasive and ELISA results can be obtained quickly and economically. The objectives were to evaluate the herd-level test characteristics of 3 commercial milk ELISA, and to determine the impact of within-herd MAP prevalence on the performance of the milk ELISA herd test. A total of 32 purposively selected herds with a median herd size of 66 milking cows were used in this 2-yr project. Fecal and milk samples were collected from all milking cows at 6-mo intervals. Fecal samples were pooled by cow age, with 5 cow samples per pool; individual fecal culture was completed on cow samples from positive pools. Herd MAP status was defined as MAP positive if, at any point during the longitudinal study, a pooled fecal culture from the herd was positive. Milk samples were analyzed using each of 3 commercial milk ELISA kits; a cow-level result from each ELISA was classified as positive following the respective manufacturer's recommended threshold for a positive result. Herd-level milk ELISA test characteristics were estimated using generalized estimating equations logistic models, which accounted for repeated measurements. Using a cutoff of 2% milk ELISA-positive cows, milk ELISA herd sensitivity relative to a herd MAP status based on all pooled fecal culture results collected during the study was as follows: ELISA A: 59% [95% confidence interval (CI): 36-78%), ELISA B: 56% (95% CI: 32-77%), and ELISA C: 63% (95% CI: 41-81%). Herd specificity for ELISA A, B, and C was 80% (95% CI: 71-88%), 96% (95% CI: 89-98%), and 92% (95% CI: 86-96%), respectively. The remainder of the analyses focused on results from ELISA B. Herd sensitivity of ELISA B increased as MAP prevalence increased. In herds with a mean MAP prevalence ≤5%, the herd sensitivity of the milk ELISA was low, ranging from 11% when MAP prevalence was 1%, to 62% when MAP prevalence was 5%. Categorical likelihood ratios based on milk ELISA within-herd prevalence predicted that herds with milk ELISA prevalence above 0 but <2% had a similar likelihood to be MAP positive or MAP negative, whereas herds with a milk ELISA prevalence between 2 and 4% were 3.7 times more likely to be MAP positive than MAP negative. All herds with a milk ELISA prevalence >4% were MAP positive. Although milk ELISA B worked well to establish herd MAP status in high-prevalence herds, interpretation was unreliable in MAP-negative and low-prevalence herds.

Laboratory evaluation of 3M Petrifilms and University of Minnesota Bi-plates as potential on-farm tests for clinical mastitis.

The objective was to determine test characteristics and compare 2 potential on-farm culture systems for clinical mastitis, the Minnesota Easy Culture System II Bi-plate and Petrifilm. The tests were evaluated using clinically positive mastitic milk samples (n = 282) to determine their ability to differentiate appropriate treatment groups; all cases that had gram-positive growth were considered treatment candidates (n = 161), whereas cases that grew gram-negative organisms only or yielded no bacterial growth were classified as no treatment (n = 121). For Petrifilm, both undiluted and 1:10 diluted milk samples were used. To create treatment categories, 2 types of Petrifilms were used, Aerobic Count (AC) and Coliform Count (CC). Both Bi-plates and Petrifilms were read after 24 h of incubation. Analysis was conducted at various colony count thresholds for the Petrifilm test system. The combination of Petrifilms that had the highest sensitivity classified a case as gram-negative if there were > or =20 colonies present on the CC. If there were <20 colonies present on the CC and >5 colonies present on the AC, a case would be classified as gram-positive. The Bi-plate had a sensitivity of 97.9% and a specificity of 68.6%. The Petrifilm test system had a sensitivity of 93.8% and a specificity of 70.1%. There was no significant difference in the sensitivities between the tests. All Bi-plates and Petrifilms were read by a laboratory technician and a group of masked readers with limited microbiology training. Kappa values for the masked readers were 0.75 for Bi-plates and 0.84 and 0.86 for AC and CC Petrifilms, respectively. The Bi-plate and Petrifilm were able to successfully categorize clinical cases of mastitis into 2 treatments based on their ability to detect the presence of a gram-positive organism. Neither method had the ability to determine if a sample was contaminated. The results of this study indicate that both tests were able to appropriately categorize cases, which could potentially result in a reduction in the quantity of antibiotics used to treat clinical cases of mastitis.

Agreement between three ELISAs for Mycobacterium avium subsp. paratuberculosis in dairy cattle.

During a 10-month period in 1999, 994 serum and tissue samples were collected from dairy cows at slaughter in eastern Canada. The sources of these cattle were from all four Atlantic Canadian provinces along with some cows from the state of Maine. The sera were used to assess the agreement of three commercially available ELISAs for Mycobacterium avium subsp. paratuberculosis. Two ELISAs were indirect absorbed ELISAs licensed for use in North America, the third was an indirect non-absorbed ELISA licensed for use in Europe. Overall, there was poor agreement between the three ELISAs. The highest and lowest kappa values were 0.33 and 0.18, which is fair and poor agreement, respectively. However, when only tissue culture-positive cattle were compared, the ELISAs had better agreement (kappa=0.37-0.51). The proportions of positive tests, however, were significantly different among the three ELISAs. The poor agreement among the three ELISAs is as concerning as the fact that these tests have low sensitivity. The implications are greatest when the tests are used at the cow level to make individual animal decisions, which is not the recommended method on the product labels. At the cow level, if the result obtained from one ELISA is positive, using a different ELISA in a pre-clinical animal has a high likelihood of giving a different result due to low predictive values of positive test results.

Comparison of two enzyme-linked immunosorbent assays for diagnosis of Mycobacterium avium subsp. paratuberculosis.

Enzyme-linked immunosorbent assays (ELISAs) are used in Johne's disease (JD) control programs as a first screening for presence of the disease in a herd. A high sensitivity of the ELISA is therefore important, yet the commonly used ELISAs have relatively low sensitivity. The inclusion of an absorption phase, although improving specificity, potentially decreases sensitivity. Sera and feces of 383 adult dairy cows in 8 herds were used to compare the test characteristics of an absorbed and a nonabsorbed indirect ELISA for the detection of JD. The absorbed ELISA is based on a protoplasmic antigen, whereas the nonabsorbed uses a lipoarabinomannan-based antigen. The potential advantage of the nonabsorbed ELISA is that it may be less specific and more sensitive. Two herds certified free of JD were used to compare the specificity of the ELISAs. The other herds used to compare sensitivity were either infected with Mycobacterium avium subsp. paratuberculosis or had unknown status. Using fecal culture as a gold standard, the diagnostic specificity for the absorbed and nonabsorbed ELISAs were 98.4% and 87.9%, respectively. The diagnostic sensitivity was 72.4% and 65.5% for the absorbed and the nonabsorbed ELISA, respectively. Furthermore, a comparison using a fecal DNA probe as the comparison standard resulted in both ELISAs having a sensitivity of 61.9%. Agreement between the 2 ELISAs was moderate, with a kappa statistic of 0.58. The nonabsorbed ELISA did not have a higher sensitivity and had a lower specificity than the absorbed ELISA. Therefore, in this population, there was no advantage gained with using the nonabsorbed ELISA.

Evaluation of three ELISAs for Mycobacterium avium subsp. paratuberculosis using tissue and fecal culture as comparison standards.

Three serum ELISAs for detection of antibodies against Mycobacterium avium subsp. paratuberculosis (Mptb) were evaluated against culture of tissue and feces samples from 994 dairy cows collected at slaughter. Culture of ileum and associated lymph nodes for Mptb were positive for 160 (16.1%) of the 994 cows and 36 (3.6%) were fecal culture-positive for Mptb. Two of the ELISAs evaluated were absorbed indirect assays and the third was a non-absorbed indirect assay. Estimated sensitivities of the absorbed ELISAs when compared to tissue culture were 8.8% and 6.9%, while the unabsorbed ELISA had a sensitivity of 16.9%. Specificities were 97.6%, 96.0% and 90.8%, respectively. When compared to fecal culture, the sensitivities of the absorbed ELISAs were 16.6% and 13.9%, respectively, and the sensitivity of the unabsorbed ELISA was 27.8%. Specificities were 97.1%, 95.9% and 90.1%, respectively. Area under the curve (AUC) of receiver operator characteristic curves for the absorbed ELISAs when tissue culture was the standard were 0.553 and 0.547, while the unabsorbed ELISA had an AUC of 0.540. When fecal culture was the comparison standard, the AUC of the absorbed ELISAs was 0.575 and 0.574, while the unabsorbed ELISA was 0.529. Overall, the sensitivities of the ELISAs when compared to tissue culture were low. The apparent advantage of the unabsorbed ELISA with respect to sensitivity is at the cost of lowered specificity and test accuracy.

Cow-level prevalence of paratuberculosis in culled dairy cows in Atlantic Canada and Maine.

The prevalence of Mycobacterium avium ssp. paratuberculosis (Mptb) in culled dairy cattle in Eastern Canada and Maine was determined to be 16.1% (95% confidence interval 13.8 to 18.3%) based on a systematic random sample of abattoir cattle. Mesenteric lymph nodes and ileum from 984 cows were examined by histologic and bacteriologic methods. Histological testing was far less sensitive than bacteriologic methods for detecting infected cattle. A seasonal pattern of positive cows was also detected, with the highest proportion of cows being Mptb-positive in June (42.5%). Overall, body condition score was not associated with prevalence of Mptb isolation.