- Email: [email protected]
Occurrence of Mycobacterium avium subsp. paratuberculosis in milk at dairy cattle farms: A systematic review and meta-analysis
Veterinary Microbiology 157 (2012) 253–263
Contents lists available at SciVerse ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Review
Occurrence of Mycobacterium avium subsp. paratuberculosis in milk at dairy cattle farms: A systematic review and meta-analysis Hisako Okura *, Nils Toft, Søren Saxmose Nielsen Department of Large Animal Sciences, Faculty of Life Sciences, University of Copenhagen, Grønnega˚rdsvej 8, DK-1870 Frederiksberg C, Denmark
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 July 2011 Received in revised form 8 December 2011 Accepted 15 December 2011
Presence of Mycobacterium avium subsp. paratuberculosis (MAP) in milk for human consumption is a concern due to its possible relationship with Crohn’s disease in humans. Pasteurization effectively reduces the MAP load by four to five logs, but the efficacy depends on the MAP concentration, which depends on the prevalence among contributing herds and individuals. Considerable variation of MAP in bulk tank milk (BTM) and individual cow’s milk (IM) is reported, but factors associated with MAP occurrence in milk at farm level have not been described. This study systematically reviewed published studies aiming at estimating the occurrence of MAP in on-farm BTM and IM by metaanalysis. A total of 692 articles were identified through electronic databases and initially screened using title and abstract. The quality of the 61 potentially relevant articles was assessed using full text and 31 articles were eventually included in the meta-analysis. The apparent prevalence (AP) of MAP in BTM and IM on farm were summarized in relation to strata defined by the test used to identify MAP and the infection status of the herds/ animals. There was considerable inconsistency in the reporting, resulting in missing information potentially explaining the dispersion in the estimated AP. The overall AP and 95% confidence intervals based on PCR and culture of MAP were summarized to 0.10 (0.04–0.22) in BTM and 0.20 (0.12–0.32) in IM. Quantifying the MAP load in test-positive milk samples was not possible because very few articles provided quantitative information on individual samples. ß 2012 Elsevier B.V. All rights reserved.
Keywords: Meta-analysis Milk Paratuberculosis Prevalence Systematic review
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . Literature search. . . . . . . . . . . 2.1. Relevance screening . . . . . . . . 2.2. 2.3. Quality assessment. . . . . . . . . Data extraction . . . . . . . . . . . . 2.4. 2.5. Data analysis . . . . . . . . . . . . . Descriptive analysis . 2.5.1. 2.5.2. Stratification . . . . . . Meta-analysis. . . . . . 2.5.3.
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
* Corresponding author. Tel.: +45 35 33 30 35; fax: +45 35 33 30 22. E-mail addresses: [email protected] (H. Okura), [email protected] (N. Toft), [email protected] (S.S. Nielsen). 0378-1135/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.vetmic.2011.12.019
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
254 254 254 254 254 254 255 255 255 255
254
3.
4.
5.
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification and description of relevant literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. 3.2. Detection limits of the test used and possible MAP load in the test-positive milk samples Apparent prevalence of MAP in BTM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Apparent prevalence of MAP in IM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Apparent prevalence of BTM and IM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Systematic review and meta-analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction Mycobacterium avium subsp. paratuberculosis (MAP) causes a chronic infection in ruminants and other animals, and can lead to significant economical losses due to reduced milk yield, premature culling, reduced body condition and eventual death (Kudahl et al., 2007). As the infection progresses, MAP is spread in the body of infected animals and the animals may shed MAP in milk (Whittington and Sergeant, 2001). Presence of MAP in milk for human consumption is of potential concern due to its possible relationship with Crohn’s disease in humans. Pasteurization and processing can be effective for four to five log reduction of the MAP load, but the initial levels of MAP may be too high for complete killing of the bacteria (Grant et al., 2005; Rademaker et al., 2007). A previous review study qualitatively and descriptively summarized information on the presence of MAP in milk without distinguishing between individual milk (IM), bulk tank milk (BTM) on farm, and raw milk at dairy processing plants (Eltholth et al., 2009). To investigate human exposure to MAP via milk, information on the quantity of MAP contamination at each level in the milk processing chain is necessary. Also knowledge about how the contributions of the MAP load at different levels in the chain impact the overall level of MAP is needed. Besides, information on possible load described in specific terms including test accuracy data may provide input to microbiological exposure assessment (Gardner, 2004). A systematic review and meta-analysis allow us to synthesize information from existing studies and also to evaluate the effect of covariates by testing heterogeneities within groups. Therefore, we systematically reviewed published studies aiming at estimating the occurrence of MAP in on-farm BTM and IM by meta-analysis, and at assessing factors associated with reported MAP variation.
2. Materials and methods The review followed a publicly available guide for systematic reviews in agri-food public health (Sargeant et al., 2005).
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
255 255 256 259 259 260 260 260 261 261
EMBASE via Ovid, ISI Web of Science and PubMed. Search keywords used were: M. avium subsp. paratuberculosis, paratuberculosis, Mycobacterium paratuberculosis, and Johne’s each combined with milk. No language restriction was imposed. In addition to the electronic databases, the proceedings of from 6th to 10th International Colloquia on Paratuberculosis from 1999 to 2009 were searched using the same keywords. The literature search included any study testing raw milk collected on farms; regardless of the specified study objectives; in order to maximize the sensitivity of the search. All identified citations with abstracts from each database were uploaded into a reference management software: RefWorks (www.refworks.com). Duplicates were removed by matching the first author, title and publication year. 2.2. Relevance screening Title and abstracts were screened for relevance by one of the authors (HO) using three inclusion criteria: (1) primary research, (2) articles in English, and (3) milk samples collected on farm and tested for presence of MAP (MAP bacteria or MAP DNA). When the abstract was not available and the relevance could not be determined from the title only, the full article was included in the quality assessment. 2.3. Quality assessment A quality assessment form was created to assess further inclusion of the articles. Full papers were obtained for all potentially relevant articles and quality was assessed by 3 reviewers (HO, NT, SSN) working independently. Disagreements among the reviewers were resolved by consensus. The criteria considered essential for inclusion in the review included: description of the milk sampling, description of the test method, reporting of the numbers of samples tested and test-positive. At this step, the articles were mapped to study level, i.e. the combination of milk samples (BTM or IM), animal species (cow or other ruminants), test used (culture, IS900 PCR, or F57 PCR) and infection status (see Section 2.5.2 for details about this stratification). Studies on animals other than cattle (i.e. sheep, goat and buffalo) were excluded.
2.1. Literature search 2.4. Data extraction Seven available electronic databases were searched for articles published between 1990 and November 2010: AGRICOLA, AGRIS, BIOSIS Reviews, CAB Abstracts and
A data extraction form was created by the authors based on the one presented in the guidelines (Sargeant et al., 2005).
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
The form was pre-tested by the 3 reviewers to ensure the consistency of the data extracted from the articles. Data extracted from the relevant studies included general information of the article, population characteristics, study design, measurement of the outcomes, statistical analysis, and results. In the review, a BTM sample referred to a sample collected from the farm bulk tank of a cow herd on a single test-day. Studies including repeated BTM samples from the same farms were excluded if the samples were obtained less than 1 year apart. The MAP presence in these samples was considered to be attributed to the herd demography, which changes considerably in 1 year. Conversely, samples taken more than 1 year apart were considered uncorrelated and thus included in the study. Studies providing only the number of BTM samples without description on the source herds (e.g. the number of the herds from which the BTM samples were collected) were excluded. An individual milk (IM) sample referred to a sample collected from an individual cow on one test-day. The study unit for IM was an individual cow. Therefore, studies describing milk samples representing each teat of the animals were excluded. Furthermore, in contrast to BTM, studies where the study unit was the test-day sample (i.e. multiple days of sampling per animal) were excluded because samples collected from one individual animal for multiple days are expected to be highly correlated given the chronic nature of the infection. 2.5. Data analysis 2.5.1. Descriptive analysis The descriptive analysis of the extracted data was done by summarizing to tables, with stratification as specified in Section 2.5.2. Some studies included in the descriptive tables were excluded in the following meta-analysis in order to keep equal weight to a set of milk samples. For example, when a set of milk samples was sent to multiple testing laboratories leading to multiple results, all were presented in the descriptive analysis; however, only one randomly selected result was used for the statistical analysis. 2.5.2. Stratification Milk samples collected on farms were characterized by 3 strata: type of milk (IM or BTM), test used for MAP detection (culture or PCR), and infection status of the herds/animals (infected or unknown). Basically two tests were used to detect MAP in milk: detection of MAP bacteria by a culture-based method or MAP DNA by PCR. There are further differences within the tests in terms of practical procedures such as preenrichment and DNA extraction but this information was not included. Detection methods, where the presence of MAP in milk included culture of MAP bacteria on solid or liquid media were grouped into ‘culture’. Detection methods, where the presence represents detection of MAP DNA specific insertion sequences, were grouped into ‘PCR’. There are essential differences within PCR methods in terms of which sequence are aimed at because there are different numbers of sequences present in a MAP cell,
255
ultimately affecting both the sensitivity and the estimated MAP load. The analytical specificity may vary depending on the PCR primers used, but there are no published data on the superiority of any of the insertion sequences used. Therefore, within the PCR group, studies were further grouped by the target insertion sequence. Studies were grouped by the infection status of the herds/animals because MAP shedding in milk depends on the infection status of the herd/animals. Herds/animals were grouped into ‘infected’ when the herds/animals were tested positive by either antibody ELISA or fecal culture prior to or simultaneously with the milk testing. Studies with no information on the infection status of the source herds were described with ‘unknown’ infection status. 2.5.3. Meta-analysis A meta-analysis using a random-effects model was used to estimate apparent prevalence (AP) in milk. Analyses were performed to summarize overall AP by BTM and IM, as well as according to the above stratification. A random-effects model was chosen because the true prevalence was assumed to vary across studies, where the reported AP reflects the true variation, random variation and differences in test characteristics (Dohoo et al., 2009, pp. 739–772). Forest plots were used to show the estimated AP of MAP in milk with corresponding 95% confidence intervals in the individual studies and based on a random-effects model (Viechtbauer, 2010). Raw proportions were logit-transformed to avoid negative confidence intervals. Cochran’s Q statistic and p-value were calculated to assess heterogeneity (Dohoo et al., 2009, pp. 739–772). The Q statistic was defined as the total dispersion in the estimated AP in which the null hypothesis tested was ‘no heterogeneity’. Furthermore, Higgins’ I2 statistic, defined as the ratio of the true heterogeneity to total variance across the observed estimate, was calculated to quantify the heterogeneity (Borenstein, 2009), because heterogeneity and dispersion due to differences in the test used and the infection status were expected. Meta-analysis of the AP of MAP in milk was done using R and the package ‘metafor’ (R Development Core Team, 2010; Viechtbauer, 2010). 3. Results 3.1. Identification and description of relevant literature Initially, 692 articles were identified through the electronic databases. After the first screening, 61 articles passed the eligibility assessment. In the quality assessment, 30 of these were excluded prior to data extraction for reasons given in Fig. 1 (Stephan et al., 2002; Stratmann et al., 2006, 2002; Caldow et al., 2007; Shankar et al., 2010; Singh et al., 2007b,c; Singh and Vihan, 2004; Barrington et al., 2003; Corti and Stephan, 2002; Grant et al., 2002, 2000; Hasonova et al., 2009; Hermon-Taylor et al., 1999; Kaittanis et al., 2007; Konuk et al., 2007; Metzger-Boddien et al., 2006; O’Reilly et al., 2004; Ronald et al., 2009; Smith et al., 2003; Taddei et al., 2008; Tasara and Stephan, 2005; Wells, 2005; DimareliMalli, 2010; Djønne et al., 2003; Giese and Ahrens, 2000; Ikonomopoulos et al., 2009;
256
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
• • •
•
• • • • •
• • •
Fig. 1. Steps and results of the systematic review and meta-analysis for Mycobacterium avium subsp. paratuberculosis in milk (bulk tank milk: BTM and individual milk: IM) at farm level.
Kumar et al., 2008; Muehlherr et al., 2003; Nebbia et al., 2006; Selvam et al., 2009; Seyyedin et al., 2008, 2010; Wilson et al., 2010; Favila-Humara et al., 2010; Dzieciol et al., 2010; Ridge et al., 2010; Herthnek et al., 2008; Pillai and Jayarao, 2002; Slana et al., 2008). Eight studies from 5 articles were excluded from the descriptive tables, and studies from 2 articles were further divided as specified in Fig. 1. As a result, the descriptive tables included 26 BTM (Table 1) and 31 IM studies (Table 2). The meta-analysis included 18 BTM and 27 IM studies, because 8 BTM and 4 IM studies were excluded due to multiple sampling (Herthnek et al., 2008; Pillai and Jayarao, 2002;
Slana et al., 2009b), lack of source herd information (Dzieciol et al., 2010; Stratmann et al., 2002) and testing at multiple labs (Ridge et al., 2010) (Tables 1 and 2, Fig. 1). 3.2. Detection limits of the test used and possible MAP load in the test-positive milk samples Information on the test used to analyze the milk samples (e.g. sensitivity, specificity, use of positive/ negative controls, detection limit) and the estimated MAP load was rarely reported. Only 19 studies from 9 articles reported either/both the detection limit and the
Table 1 Apparent prevalence of Mycobacterium avium subsp. paratuberculosis (MAP) in bulk tank milk on cattle farms. Test used
Number of samples tested
Number of samples positive
Apparent prevalence
Country
Study period
Authors
FC positive Mixed*3 Infected Infected Infected ELISA positive ELISA positive Unknown Unknown Unknown Unknown
20 29 91 91 91 14 52 69 237 220 225
1 2 2 0 0 10 0 5 4 0 2
0.05 0.07 0.02 0.00 0.00 0.71 0.00 0.07 0.02 0.00 0.01
USA USA Australia Australia Australia Central Mexico USA Iran Czech Republic Cyprus Cyprus
NR*1 1999–2000 2006 2006 2006 NR NR 2006–2007 NR 2007 2007–2008
Pillai and Jayarao (2002)*5,17 Jayarao et al. (2004) Ridge et al. (2010)*16 Ridge et al. (2010)*16 Ridge et al. (2010)*16 Favila-Humara et al. (2010) Stabel et al. (2002) Seyyedin et al. (2010) Slana et al. (2009a) Slana et al. (2009b) Botsaris et al. (2010)
FC positive Mixed*3 ELISA positive Infected*2 Infected Mixed*4 Unknown Unknown Unknown Unknown Unknown
20 29 14 52 5 143 16 110 237 220 225
10 8 10 35 4 19 0 12 82 49 50
0.5 0.28 0.71 0.67 0.80 0.13 0.00 0.11 0.35 0.22 0.22
USA USA Central Mexico USA Argentina Denmark Brazil Iran Czech Republic Cyprus Cyprus
NR 1999–2002 NR 2007 2006 NR 2006 NR 2007 2007–2008
Pillai and Jayarao (2002)*7,17 Jayarao et al. (2004) Favila-Humara et al. (2010) Stabel et al. (2002) Slana et al. (2008)*8,17 Herthnek et al. (2008)*6,13,17 Carvalho et al. (2009) Haghkhah et al. (2008) Slana et al. (2009a) Slana et al. (2009b)*6,14 Botsaris et al. (2010)*9
Unknown Unknown Infected
100 220 5
3 14 2
0.03 0.06 0.4
Switzerland Cyprus Argentina
2005 2007 2007
Bosshard et al. (2006)*10 Slana et al. (2009b)*6 Slana et al. (2008)*11,17
7
5
0.71
Germany
NR
Stratmann et al. (2002)*12,15,18
ELISA positive
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Culture-based methods (n = 11) HEYM HEYM BACTEC BACTEC BACTEC HEYM and/or Lowenstein-Jensen HEYM HEYM HEYM HEYM HEYM IS900 PCR-based methods (n = 11) IS900 PCR IS900 PCR IS900 PCR IS900 PCR IS900 PCR IS900 realtime PCR IS900 PCR IS900 nested PCR IS900 quantitative PCR IS900 quantitative PCR IS900 combined phage PCR F57 PCR-based method (n = 3) F57 light cycler-based real-time PCR F57 qPCR F57 realtimePCR Other PCR-based method (n = 1) Phage and peptide-mediated capture ISMAv2 PCR
Infection status of the herd
*1 NR: Not reported. *2: No information on the test used for the classification. *3: 16/29 herds were ELISA and/or FC positive. *4: Environmentally positive and negative. Reported detection limit of the test used: *5: 10–100 cfu/ml milk; *6: 100 organisms/ml; *7: 10 cfu/ml; *8: 5–6 MAP cells per/ml; *9: single units of organisms/ml; *10: 100 cells/ml; *11: 83 MAP cells/ml; *12: 10 MAP/ml. Reported estimated MAP load in the positive samples: *13: rarely exceed 100 organisms/ml; *14: single units to tens of organisms/ml; *15: rarely exceed 100 organisms/ml. *16: The set of samples were sent to 3 laboratories. One randomly selected study was used in the meta-analysis. *17: Excluded from the meta-analysis due to multiple sampling. *18: Excluded from the meta-analysis due to lack of source herd information.
257
258
Table 2 Apparent prevalence of Mycobacterium avium subsp. paratuberculosis (MAP) in individual milk samples. Test used
Culture-based methods (n = 13) HEYM
Infection status of the animals
Number of Number of Apparent Country prevalence samples samples positive tested
Study period
Authors
46
0.35
Canada
NR*1
Gao et al. (2009)*2,7
11 24 211 12 87 26 50 1493 126 14 15 15
6 2 9 7 2 25 42 43 3 4 4 1
0.55 0.08 0.04 0.58 0.02 0.96 0.84 0.03 0.02 0.29 0.27 0.07
Denmark Argentina USA Iran Poland India India USA USA Australia Australia Australia
NR NR NR 2006–2007 NR 2004 NR 1999–2002 NR 2006 2006 2006
Giese and Ahrens (2000) Paolicchi et al. (2003) Pillai and Jayarao (2002)*2,8 Seyyedin et al. (2010) Szteyn et al. (2008) Singh et al. (2007a,b,c) Sharma et al. (2008) Jayarao et al. (2004) Streeter et al. (1995) Ridge et al. (2010)*9 Ridge et al. (2010)*9 Ridge et al. (2010)*9
134
71
0.54
Canada
NR
Gao et al. (2009)*3
11
2
0.18
Denmark NR
Giese and Ahrens (2000)
Nested IS900 PCR 11 IS900 PCR 103 IS900 PCR 24 IS900 PCR 211 IS900 PCR 206 IS900 PCR 56 IS900 PCR 87 IS900 PCR 50 IS900 PCR 1493 IS900 PCR 9 IS900 nested PCR 328 Nested IS900 PCR 98 Nested IS900 PCR 46 IS900 realtime PCR 342 Other PCR-based method (n = 2) F57 realtime PCR Unknown 342 Phage and peptide-mediated capture ISMAv2 PCR Serologically positive for MAP in milk or blood ELISA 9
3 21 0 69 8 10 18 3 201 3 13 17 9 64
0.27 0.20 0.00 0.33 0.04 0.18 0.21 0.06 0.13 0.33 0.04 0.17 0.20 0.19
USA Poland Argentina USA Brazil Iran Poland India USA India USA USA USA Argentina
Buergelt and Williams (2004) Wiszniewska-Laszczych et al. (2009) Paolicchi et al. (2003) Pillai and Jayarao (2002)*4 Carvalho et al. (2009) Soltani et al. (2008) Szteyn et al. (2008) Sharma et al. (2008) Jayarao et al. (2004) Kaur et al. (2010) Pinedo et al. (2008a) Pinedo et al. (2008b) Buergelt and Williams (2004) Slana et al. (2008)*5,10
47 7
0.14 0.78
Argentina 2007 Germany NR
LJ and Dubos broth HEYM. HEYM HEYM HEYM HEYM HEYM HEYM HEYM Radiometric culture Radiometric culture Radiometric culture IS900 PCR-based methods (n = 16) Nested IS900 PCR IS900 PCR
Positive on FC or milk or serum ELISA Positive on FC or milk or serum ELISA Positive on serum ELISA Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Subclinical Unknown
Abbreviations: NR, not reported; FC, fecal culture; LJ, Lo¨wenstein–Jensen medium; HEYM, Herrold’s Egg Yolk Medium. Reported detection limits: *2: 10–100 colony forming units (cfu)/ml milk; *3: 2 cfu/ml; *4: 10 cfu/ml; *5: 5–6 MAP cells/ml; *6: 10 MAP/ml. Reported estimated MAP load in the positive samples: *7: 1–5 cfu/ml; *8: 10–100 cfu/ml. *9: The set of samples were sent to 3 laboratories. One randomly selected study was used in the meta-analysis. *10: Excluded from the meta-analysis due to multiple sampling.
NR NR NR NR NR NR NR NR 1999–2002 NR*1 2004–2006 2003–2004 NR 2007
Slana et al. (2008)*5,10 Stratmann et al. (2002)*6
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
133
Positive on FC or milk or serum ELISA FC positive Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
estimated MAP load (Tables 1 and 2). Hence, the estimated MAP load could not be summarized due to different types of milk samples and tests used. Detection limit by culture was reported in one article to be 10–100 cfu/ml milk (Pillai and Jayarao, 2002). Detection limits of IS900, F57 and ISMAv2 PCR were reported in 7 articles ranging from single units to 100 MAP cells/ml of milk (Bosshard et al., 2006; Botsaris et al., 2010; Gao et al., 2009; Herthnek et al., 2008; Pillai and Jayarao, 2002; Slana et al., 2008; Stratmann et al., 2002). Estimated MAP load in test-positive samples was reported in 5 articles, 3 for BTM and 2 for IM. The 3 articles reported estimated MAP load in BTM samples by either IS900 or F57 PCR to be less than 100 MAP organisms/ml of milk (Bosshard et al., 2006; Herthnek et al., 2008; Slana et al., 2009b), while no estimated MAP load by culture was reported. Estimated MAP load in IM samples was reported to be 10–100 cfu/ml milk by culture (Pillai and Jayarao, 2002) and 1–5 cfu/ml milk by IS900 PCR (Gao et al., 2009). 3.3. Apparent prevalence of MAP in BTM The AP among the BTM samples ranged between 0 and 71%. Most reported AP based on culture were below 0.1; however, one study had very high AP of 0.71 (95% confidence interval (C.I.) 0.48–0.95). The AP from PCR methods varied considerably among the studies. The AP from F57 PCR was significantly lower than that of IS900 PCR (Fig. 2). For 18 cow BTM studies, summarized AP and 95% C.I. were 0.10 (0.01–0.13). Results of the subgroup analyses for culture, IS900 PCR and F57 PCR were 0.03 (0.01–0.13),
259
0.30 (0.16–0.49), and 0.05 (0.03–0.1), respectively. The results of further stratification based on the infection status of the herds within the culture and IS900 PCR are shown in Table 3. For the culture group, there was one study in the infected stratum (positive by ELISA or fecal culture) group with very a high prevalence, which resulted in a higher AP with much wider confidence interval. For the IS900 PCR group, the AP for the infected herd group was significantly higher than the unknown herds. There were considerable heterogeneities for the overall meta-analysis and the following sub-group analyses. However, reduction in the I2 statistics was observed as the studies were grouped into the different strata (Table 3). 3.4. Apparent prevalence of MAP in IM Thirty-one studies were included in the descriptive table (Table 2). Many studies reported infection status of the herd from which the individual cows were sampled, while 4 out of 31 studies reported infection status of the animals from which the sample were collected. The AP of IM samples ranged between 0 and 96%. Overall, for 27 IM studies, the summarized AP of MAP from the meta-analysis was 0.20 (0.12–0.32) (Fig. 3, Table 3). Results of the subgroup analysis for the estimated AP based on the culture and IS900 PCR were 0.23 (0.07–0.54) and 0.16 (0.11–0.25), respectively (Fig. 3, Table 3). Stratification by infection status of the cows showed higher AP for ELISA or FC positive cows than for unknown infection status cows.
Fig. 2. Apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis in bulk tank milk on cattle farms.
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
260
Table 3 Summary of the meta-analysis of apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis (MAP) in bulk tank milk (BTM) and individual milk (IM), stratified by diagnostic test and infection status. Stratification BTM Overall Culture Known infected Unknown status IS900 PCR Known infected Unknown F57 PCR IM Overall Culture Known infected Unknown status IS900 PCR Known infected Unknown status
Estimated AP
95% C.I.
I2
Q (p-value)
18 8 3 5 8 2 6 2
0.10 0.03 0.06 0.01 0.30 0.68 0.22 0.05
0.04, 0.01, 0.00, 0.00, 0.16, 0.56, 0.15, 0.03,
0.22 0.13 0.77 0.08 0.49 0.78 0.31 0.10
97.11 89.12 89.86 70.69 95.12 0.00 83.53 32.25
<0.0001 <0.0001 <0.0001 0.01 <0.0001 0.01 <0.0001 0.22
27a 11 2 9 15 3 12
0.20 0.23 0.39 0.20 0.16 0.36 0.14
0.12, 0.07, 0.24, 0.05, 0.11, 0.17, 0.09,
0.32 0.54 0.57 0.56 0.25 0.61 0.21
97.15 97.39 40.47 97.01 93.52 65.54 91.99
<0.0001 <0.0001 0.20 <0.0001 <0.0001 <0.0001 <0.0001
Number of studies
a One study tested the milk samples using phage and peptide-mediated capture ISMAv2 PCR was included in the overall AP estimate but not in the subgroup analysis.
There were considerable heterogeneities for the overall meta-analysis and the following sub-group analyses. Reduction in the I2 statistics as the studies were grouped by the characteristics was only observed in the positive cows and considerable heterogeneity remained in the unknown infection status cow group (Table 3). 4. Discussion
infection status tended to be lower than the AP among known infected herds/animals which was not surprising. It should not be interpreted as if BTMs and IMs without information on infection status have lower prevalence of MAP, but rather that the information on the infection status (test history) could be used to reduce MAP positivity in milk. Summarized AP for BTM was lower than that of IMs which indicated a dilution effect on BTM. It was not possible to quantify the possible MAP load in milk on farm.
4.1. Apparent prevalence of BTM and IM 4.2. Systematic review and meta-analysis This systematic review and meta-analysis estimated the probability of detecting MAP in milk collected on farms. For BTM, AP from culture was lower than AP from IS900 PCR, while in IM AP from culture was higher than AP from IS900 PCR. There are two likely explanations for the different AP in BTM and IM by culture and IS900 PCR: (1) sensitivity of PCR is higher than culture due to a lower detection limit of PCR; or (2) PCR also detects dead MAP. These two circumstances would have different implications at herd and animal level. An individual excreting MAP in milk would probably excrete a relatively high number of viable bacteria, and the effect of dead MAP would be minor. Consequently, the diagnostic test would not be greatly affected by low analytical sensitivity. Conversely, the concentration in BTM is highly affected by the within-herd prevalence of MAP shedders. The high AP for IS900 PCR may reflect higher analytical sensitivity compared to culture. Quantification of the MAP load in the samples would be needed to resolve this issue. In the sub-group analysis for the infection status, summary AP for both BTM and IM were significantly higher for the infected herds/animals than unknown herds/animals (Table 3). It should be noted that there tends to be larger number of samples within the study for unknown herds/animals and caution is needed when interpreting the results from a small number of studies. ‘Unknown’ infection status can include infected herds/animals. AP among herds/animals of unknown
Our search strategy was not limited to prevalence studies. Due to inclusion of non-prevalence studies, the review had to compromise on the quality of the study designs. Therefore many included studies did not adequately describe the choice of sampling scheme, which could have introduced reporting bias. Even taking the reporting bias into consideration, the overall quality of reporting was poor. Many studies did not report information of the test used, e.g. sensitivity and specificity, and even animal species was not clear in two articles (Dzieciol et al., 2010; Stratmann et al., 2002). Meta-analysis in observational studies is less common than randomized control trials, partly because standardized methods of reporting observational studies are relatively new (von Elm et al., 2007; Gardner et al., 2011). Still, there are studies with good quality of reporting (Gao et al., 2009; Herthnek et al., 2008). Heterogeneities in a meta-analysis can occur due to differences in many factors such as country, breed, on-farm management, test used, and definition of the outcomes. These diversities are referred to as ‘clinical’ or ‘methodological’ heterogeneities and may or may not be responsible for the summary estimate. Clinical heterogeneity is a variation explained by the biological context, while statistical heterogeneity exists when the AP truly differs across the studies (Higgins and Green, 2011; Higgins and Thompson, 2002). A heterogeneity indicator such as
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
261
Fig. 3. Apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis in milk from individual cows.
Cochran’s Q statistic does not distinguish between different sources of heterogeneity. Hence, it is often difficult to draw conclusions from a meta-analysis when clinical heterogeneity is expected in the summary estimate. A sub-group analysis is one way to investigate the heterogeneity when clinical differences are expected (Barrington et al., 2003; Dohoo et al., 2009, pp. 739–772). If the milk samples from the same infection statuses of animals/herds are tested by the same test the resulting AP should not vary, although there is a variation due to laboratories. Sub-group analysis by diagnostic test and infection status indicated that because the heterogeneity became smaller these subgroups could partly explain the heterogeneity among the studies. Thus, our summary AP and the similarity and differences among the group and sub-groups could reflect the real situation in the probability of detecting MAP in milk on farm.
5. Conclusion The reported AP and 95% C.I. based on PCR and culture of MAP were summarized to 0.10 (0.04–0.2) in BTM and 0.20 (0.12–0.32) in IM. MAP prevalence in milk from antibody ELISA or fecal culture positive herds/animals seemed to be higher than herds/animals with unknown infection status. Considerable variation was observed among the studies, and the MAP load could not be estimated. The quality of the reporting varied considerably resulting in missing information potentially explaining the dispersion in the estimated AP.
References Barrington, G.M., Gay, J.M., Eriks, I.S., Davis, W.C., Evermann, J.F., Emerson, C., O‘Rourke, J.L., Hamilton, M.J., Bradway, D.S., 2003. Temporal patterns of diagnostic results in serial samples from cattle with advanced paratuberculosis infections. J. Vet. Diagn. Invest. 15, 195–200. Borenstein, M., 2009. Introduction to Meta-analysis. Wiley-Blackwell, Chichester, United Kingdom. Bosshard, C., Stephan, R., Tasara, T., 2006. Application of an F57 sequencebased real-time PCR assay for Mycobacterium paratuberculosis detection in bulk tank raw milk and slaughtered healthy dairy cows. J. Food Prot. 69, 1662–1667. Botsaris, G., Slana, I., Liapi, M., Dodd, C., Economides, C., Rees, C., Pavlik, I., 2010. Rapid detection methods for viable Mycobacterium avium subspecies paratuberculosis in milk and cheese. Int. J. Food Microbiol. 141 (Suppl. 1), S87–S90. Buergelt, C.D., Williams, J.E., 2004. Nested PCR on blood and milk for the detection of Mycobacterium avium subsp paratuberculosis DNA in clinical and subclinical bovine paratuberculosis. Aust. Vet. J. 82, 497–503. Caldow, G., Strain, S.A.J., Chapman, Z., Kemp, R., Cook, A.J., 2007. A survey to estimate the herd level prevalence of paratuberculosis in the dairy herd of the United Kingdom. Cattle Pract. 15, 169–171. Carvalho, I.A., Silva Jr., A., Campos, V.E., Moreira, M.A., 2009. Short communication: detection of Mycobacterium avium subspecies paratuberculosis by polymerase chain reaction in bovine milk in Brazil. J. Dairy Sci. 92, 5408–5410. Corti, S., Stephan, R., 2002. Detection of Mycobacterium avium subspecies paratuberculosis specific IS900 insertion sequences in bulk-tank milk samples obtained from different regions throughout Switzerland. BMC Microbiol. 2, 15. DimareliMalli, Z., 2010. Detection of Mycobacterium avium subsp. paratuberculosis in milk from clinically affected sheep and goats. Int. J. Appl. Res. Vet. Med. 8, 44–50. Djønne, B., Jensen, M.R., Grant, I.R., Holstad, G., 2003. Detection by immunomagnetic PCR of Mycobacterium avium subsp. paratuberculosis in milk from dairy goats in Norway. Vet. Microbiol. 92, 135–143. Dohoo, I., Martin, W., Stryhn, H., 2009. Veterinary Epidemiologic Research. VER Inc., Charlottetown, Canada, pp. 739–772.
262
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Dzieciol, M., Volgger, P., Khol, J., Baumgartner, W., Wagner, M., Hein, I., 2010. A novel real-time PCR assay for specific detection and quantification of Mycobacterium avium subsp, paratuberculosis in milk with the inherent possibility of differentiation between viable and dead cells. BMC Res. Notes 3, 251. Eltholth, M.M., Marsh, V.R., Van Winden, S., Guitian, F.J., 2009. contamination of food products with Mycobacterium avium paratuberculosis: a systematic review. J. Appl. Microbiol. 107, 1061–1071. Favila-Humara, L.C., Chavez-Gris, G.G., Carrillo-Casas, E.M., HernandezCastro, R., 2010. Mycobacterium avium subsp. paratuberculosis detection in individual and bulk tank milk samples from bovine herds and caprine flocks. Foodborne Pathog. Dis. 7, 351–355. Gao, A., Odumeru, J., Raymond, M., Hendrick, S., Duffield, T., Mutharia, L., 2009. Comparison of milk culture, direct and nested polymerase chain reaction (PCR) with fecal culture based on samples from dairy herds infected with Mycobacterium avium subsp. paratuberculosis. Can. J. Vet. Res. 73, 58–64. Gardner, I.A., 2004. An epidemiologic critique of current microbial risk assessment practices: the importance of prevalence and test accuracy data. J. Food Prot. 67, 2000–2007. Gardner, I.A., Nielsen, S.S., Whittington, R.J., Collins, M.T., Bakker, D., Harris, B., Sreevatsan, S., Lombard, J.E., Sweeney, R., Smith, D.R., Gavalchin, J., Eda, S., 2011. Consensus-based reporting standards for diagnostic test accuracy studies for paratuberculosis in ruminants. Prev. Vet. Med. 101, 18–34. Giese, S.B., Ahrens, P., 2000. Detection of Mycobacterium avium subsp. paratuberculosis in milk from clinically affected cows by PCR and culture. Vet. Microbiol. 77, 291–297. Grant, I.R., Hitchings, E.I., McCartney, A., Ferguson, F., Rowe, M.T., 2002. Effect of commercial-scale high-temperature, short-time pasteurization on the viability of Mycobacterium paratuberculosis in naturally infected cows’ milk. Appl. Environ. Microbiol. 68, 602–607. Grant, I.R., Pope, C.M., O‘Riordan, L.M., Ball, H.J., Rowe, M.T., 2000. Improved detection of Mycobacterium avium subsp. paratuberculosis in milk by immunomagnetic PCR. Vet. Microbiol. 77, 369–378. Grant, I.R., Williams, A.G., Rowe, M.T., Muir, D.D., 2005. Efficacy of various pasteurization time-temperature conditions in combination with homogenization on inactivation of Mycobacterium avium subsp. paratuberculosis in milk. Appl. Environ. Microbiol. 71, 2853–2861. Haghkhah, M., Ansari-Lari, M., Novin-Baheran, A.M., Bahramy, A., 2008. Herd-level prevalence of Mycobacterium avium subspecies paratuberculosis by bulk-tank milk PCR in Fars province (Southern Iran) Dairy Herds. Prev. Vet. Med. 86, 8–13. Hasonova, L., Trcka, I., Babak, V., Rozsypalova, Z., Pribylova, R., Pavlik, I., 2009. Distribution of Mycobacterium avium subsp. paratuberculosis in tissues of naturally infected cattle as affected by age. Vet. Med. 54, 257–269. Hermon-Taylor, J., Cheng, J., Bull, T., 1999. Mycobacterium avium subsp. paratuberculosis in milk: an animal and human pathogen (meeting abstract). J. Microbiol. Methods 38, 217. Herthnek, D., Nielsen, S.S., Lindberg, A., Bolske, G., 2008. A robust method for bacterial lysis and DNA purification to be used with real-time PCR for detection of Mycobacterium avium subsp. paratuberculosis in milk. J. Microbiol. Methods 75, 335–340. Higgins, J.P., Green, S., 2011. Cochrane handbook for systematic reviews of interventions version 5.1.0 [updated March 2011]. The Cochrane Collaboration. Available at: http://www.cochrane.org/training/ cochrane-handbook. Higgins, J.P., Thompson, S.G., 2002. Quantifying heterogeneity in a metaanalysis. Stat. Med. 21, 1539–1558. Ikonomopoulos, J., Liandris, E., Andreadou, M., Gazouli, M., 2009. Associations between abortion-records in goats and test-positivity to Mycobacterium avium subsp. paratuberculosis. Open Vet. Sci. J. 3, 1–5. Jayarao, B.M., Pillai, S.R., Wolfgang, D.R., Griswold, D.R., Rossiter, C.A., Tewari, D., Burns, C.M., Hutchinson, L.J., 2004. Evaluation of IS900-PCR assay for detection of Mycobacterium avium subspecies paratuberculosis infection in cattle using quarter milk and bulk tank milk samples. Foodborne Pathog. Dis. 1, 17–26. Kaittanis, C., Naser, S.A., Perez, J.M., 2007. One-step, nanoparticlemediated bacterial detection with magnetic relaxation. Nano Lett. 7, 380–383. Kaur, P., Filia, G., Singh, S.V., Patil, P.K., Sandhu, K.S., 2010. Molecular detection and typing of Mycobacterium avium subspecies paratuberculosis from milk samples of dairy animals. Trop. Anim. Health Prod. 42, 1031–1035. Konuk, M., Korcan, E., Dulgerbaki, S., Altindis, M., 2007. Isolation and identification of Mycobacteria from raw milk samples in Afyonkarahisar District of Turkey. Int. J. Food Microbiol. 115, 343–347. Kudahl, A.B., Sorensen, J.T., Nielsen, S.S., Ostergaard, S., 2007. Simulated economic effects of improving the sensitivity of a diagnostic test in paratuberculosis control. Prev. Vet. Med. 78, 118–129.
Kumar, S., Singh, S.V., Sevilla, I., Singh, A.V., Whittington, R.J., Juste, R.A., Sharma, G., Singh, P.K., Sohal, J.S., 2008. Lacto-prevalence, genotyping of Mycobacterium avium subspecies paratuberculosis and evaluation of three diagnostic tests in milk of naturally infected goatherds. Small Ruminant Res. 74, 37–44. Metzger-Boddien, C., Khaschabi, D., Schonbauer, M., Boddien, S., Schlederer, T., Kehle, J., 2006. Automated high-throughput immunomagnetic separation-PCR for detection of Mycobacterium avium subsp. paratuberculosis in bovine milk. Int. J. Food Microbiol. 110, 201–208. Muehlherr, J.E., Zweifel, C., Corti, S., Blanco, J.E., Stephan, R., 2003. Microbiological quality of raw goat’s and ewe’s bulk-tank milk in Switzerland. J. Dairy Sci. 86, 3849–3856. Nebbia, P., Robino, P., Zoppi, S., De Meneghi, D., 2006. Detection and excretion pattern of Mycobacterium avium subspecies paratuberculosis in milk of asymptomatic sheep and goats by nested-PCR. Small Ruminant Res. 66, 116–120. O’Reilly, C.E., O’Connor, L., Anderson, W., Harvey, P., Grant, I.R., Donaghy, J., Rowe, M., O’Mahony, P., 2004. Surveillance of bulk raw and commercially pasteurized cows’ milk from approved Irish liquid-milk pasteurization plants to determine the incidence of Mycobacterium paratuberculosis. Appl. Environ. Microbiol. 70, 5138–5144. Paolicchi, F.A., Zumarraga, M.J., Gioffre, A., Zamorano, P., Morsella, C., Verna, A., Cataldi, A., Alito, A., Romano, M., 2003. Application of different methods for the diagnosis of paratuberculosis in a dairy cattle herd in Argentina. J. Vet. Med. B: Infect. Dis. Vet. Public Health 50, 20–26. Pillai, S.R., Jayarao, B.M., 2002. Application of IS900 PCR for detection of Mycobacterium avium subsp. paratuberculosis directly from raw milk. J. Dairy Sci. 85, 1052–1057. Pinedo, P.J., Rae, D.O., Williams, J.E., Donovan, G.A., Melendez, P., Buergelt, C.D., 2008a. Association among results of serum ELISA, faecal culture and nested PCR on milk, blood and faeces for the detection of paratuberculosis in dairy cows. Transbound Emerg. Dis. 55, 125–133. Pinedo, P.J., Williams, J.E., Monif, G.R.G., Rae, D.O., Buergelt, C.D., 2008b. Mycobacterium paratuberculosis shedding into milk: association of ELISA seroreactivity with DNA detection in milk. Int. J. Appl. Res. Vet. Med. 6, 137–144. R Development Core Team, 2010. R: A Language and Environment for Statistical Computing. Rademaker, J.L., Vissers, M.M., Te Giffel, M.C., 2007. Effective heat inactivation of Mycobacterium avium subsp. paratuberculosis in raw milk contaminated with naturally infected feces. Appl. Environ. Microbiol. 73, 4185–4190. Ridge, S.E., Andreata, S., Jones, K., Cantlon, K., Francis, B., Florisson, N., Gwozdz, J., 2010. Inter-laboratory comparison of radiometric culture for Mycobacterium avium subsp. paratuberculosis using raw milk from known infected herds and individual dairy cattle in Victoria. Aust. Vet. J. 88, 249–254. Ronald, B.S.M., Narayanan, R., Krishnakumar, N., Baegan, S., Manohar, B.M., 2009. Detection of Mycobacterium avium subsp. paratuberculosis excretion in goat milk. Asian J. Microbiol. Biotechnol. Environ. Sci. 11, 457–460. Sargeant, J.M., Amezcua, M.D.R., Rajic, A., Waddell, L., 2005. A Guide to Conducting Systematic Reviews in Agri-Food Public Health. Food Safety Research and Response Network. Selvam, A., Vijayarani, K., Kumanan, K., 2009. Detection of Mycobacterium avium subsp paratuberculosis in milk of asymptomatic sheep by polymerase chain reaction. Indian J. Anim. Sci. 79, 784–785. Seyyedin, M., Salehi, T.Z., Tadjbakhsh, H., Najafi, M.F., Rabbani, M., 2008. Isolation and identification of Mycobacterium avium subsp. paratuberculosis from milk, manure and fecal samples of holstein-friesian dairy cattle. Pakistan J. Biol. Sci. 11, 2639–2645. Seyyedin, M., Zahraei, T., Najafi, M.F., 2010. Comparison of isolation frequency of Mycobacterium avium subspecies paratuberculosis from different types of samples. Pakistan Vet. J. 30, 143–149. Shankar, H., Singh, S.V., Singh, P.K., Singh, A.V., Sohal, J.S., Greenstein, R.J., 2010. Presence, characterization, and genotype profiles of Mycobacterium avium subspecies paratuberculosis from unpasteurized individual and pooled milk, commercial pasteurized milk, and milk products in India by culture, PCR, and PCR-REA methods. Int. J. Infect. Dis. 14, e121–e126. Sharma, G., Singh, S.V., Sevilla, I., Singh, A.V., Whittington, R.J., Juste, R.A., Kumar, S., Gupta, V.K., Singh, P.K., Sohal, J.S., Vihan, V.S., 2008. Evaluation of indigenous milk ELISA with m-Culture and m-PCR for the diagnosis of bovine Johne’s disease (BJD) in lactating Indian dairy cattle. Res. Vet. Sci. 84, 30–37. Singh, S.V., Vihan, V.S., 2004. Detection of Mycobacterium avium subspecies paratuberculosis in goat milk. Small Ruminant Res. 54, 231–235. Singh, S.V., Singh, A.V., Singh, R., Sandhu, K.S., Singh, P.K., Sohal, J.S., Gupta, V.K., Vihan, V.S., 2007a. Evaluation of highly sensitive
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263 indigenous milk ELISA Kit with fecal culture, milk culture and fecalPCR for the diagnosis of bovine Johne’s disease (BJD) in India. Comp. Immunol. Microbiol. Infect. Dis. 30, 175–186. Singh, S.V., Singh, A.V., Singh, R., Sandhu, K.S., Singh, P.K., Sohal, J.S., 2007b. Comparison of multiple diagnostic tests for the diagnosis of bovine paratuberculosis in lactating cows. Indian J. Anim. Sci. 77, 1115–1117. Singh, S.V., Singh, P.K., Singh, A.V., Sohal, J.S., Subodh, S., Narayanasamy, K., 2007c. Non-chemical method of DNA recovery and characterization of Mycobacterium avium subspecies paratuberculosis using IS900 PCR. Indian J. Exp. Biol. 45, 812–816. Slana, I., Kralik, P., Kralova, A., Pavlik, I., 2008. On-farm spread of Mycobacterium avium subsp. paratuberculosis in raw milk studied by IS900 and F57 competitive real time quantitative PCR and culture examination. Int. J. Food Microbiol. 128, 250–257. Slana, I., Bartos, M., Roubal, P., Babak, V., Pavlik, I., 2009a. Mycobacterium avium subsp. paratuberculosis and M. a. avium detected by culture, IS900 and IS901 highly sensitive PCR in Bulk tank milk from dairy herds in the Czech Republic between 2002 and 2004. Czech J. Food Sci. 27, 372–378. Slana, I., Liapi, M., Moravkova, M., Kralova, A., Pavlik, I., 2009b. Mycobacterium avium subsp. paratuberculosis in cow bulk tank milk in Cyprus detected by culture and quantitative IS900 and F57 realtime PCR. Prev. Vet. Med. 89, 223–226. Smith, W.L., McGarvey, K.L., Cullor, J.S., 2003. The use of spiral plating and microscopic colony counting for the rapid quantitation of Mycobacterium paratuberculosis. Lett. Appl. Microbiol. 36, 293–296. Soltani, M., Nassiry, M.R., Shahroudi, F.E., Bassam, M.R., 2008. Detection of Mycobacterium paratuberculosis in feces and milk samples from holstein dairy cows by PCR. Biotechnology 7, 582–585. Stabel, J.R., Wells, S.J., Wagner, B.A., 2002. Relationships between fecal culture, ELISA, and bulk tank milk test results for Johne’s disease in US dairy herds. J. Dairy Sci. 85, 525–531. Stephan, R., Buhler, K., Corti, S., 2002. Incidence of Mycobacterium avium subspecies paratuberculosis in bulk-tank milk samples from different regions in Switzerland. Vet. Rec. 150, 214–215. Stratmann, J., Dohmann, K., Heinzmann, J., Gerlach, G.F., 2006. Peptide aMptD-mediated capture PCR for detection of Mycobacterium avium
263
subsp. paratuberculosis in bulk milk samples. Appl. Environ. Microbiol. 72, 5150–5158. Stratmann, J., Strommenger, B., Stevenson, K., Gerlach, G.F., 2002. Development of a peptide-mediated capture PCR for detection of Mycobacterium avium subsp. paratuberculosis in milk. J. Clin. Microbiol. 40, 4244–4250. Streeter, R.N., Hoffsis, G.F., Bech-Nielsen, S., Shulaw, W.P., Rings, D.M., 1995. Isolation of Mycobacterium paratuberculosis from colostrum and milk of subclinically infected cows. Am. J. Vet. Res. 56, 1322–1324. Szteyn, J., Wiszniewska-Laszczych, A., Ruszczynska, A., 2008. Effectiveness of Mycobacterium paratuberculosis isolation from raw milk by means of direct isolation of DNA and classic culture. Pol. J. Vet. Sci. 11, 25–28. Taddei, R., Barbieri, I., Pacciarini, M.L., Fallacara, F., Belletti, G.L., Arrigoni, N., 2008. Mycobacterium porcinum strains isolated from bovine bulk milk: implications for Mycobacterium avium subsp. paratuberculosis detection by PCR and culture. Vet. Microbiol. 130, 338–347. Tasara, T., Stephan, R., 2005. Development of an F57 sequence-based realtime PCR assay for detection of Mycobacterium avium subsp. paratuberculosis in milk. Appl. Environ. Microbiol. 71, 5957–5968. Viechtbauer, W., 2010. Conducting meta-analyses in R with the metafor package. J. Stat. Softw. 36, 1. von Elm, E., Altman, D.G., Egger, M., Pocock, S.J., Gotzsche, P.C., Vandenbroucke, J.P., STROBE Initiative, 2007. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 370, 1453–1457. Wells, S.J., 2005. Is milk or feces more important in transmission of map? In: Proceedings of the North American Veterinary Conference, Large Animal, vol. 19. p. 55. Whittington, R.J., Sergeant, E.S., 2001. Progress Towards understanding the spread, detection and control of Mycobacterium avium subsp paratuberculosis in animal populations. Aust. Vet. J. 79, 267–278. Wilson, D.J., Rood, K., Biswas, P., Byrem, T.M., 2010. Herd-level prevalence of Johne’s disease in Utah and adjacent areas of the intermountain West as detected by a bulk-tank milk surveillance project. J. Dairy Sci. 93, 5792–5797. Wiszniewska-Laszczych, A., Szteyn, J., Smolinska, A., 2009. Analysis of correlations between the occurrence of anti-MAP antibodies in blood serum and the presence of DNA-MAP in milk. Pol. J. Vet. Sci. 12, 379–383.
Contents lists available at SciVerse ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Review
Occurrence of Mycobacterium avium subsp. paratuberculosis in milk at dairy cattle farms: A systematic review and meta-analysis Hisako Okura *, Nils Toft, Søren Saxmose Nielsen Department of Large Animal Sciences, Faculty of Life Sciences, University of Copenhagen, Grønnega˚rdsvej 8, DK-1870 Frederiksberg C, Denmark
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 July 2011 Received in revised form 8 December 2011 Accepted 15 December 2011
Presence of Mycobacterium avium subsp. paratuberculosis (MAP) in milk for human consumption is a concern due to its possible relationship with Crohn’s disease in humans. Pasteurization effectively reduces the MAP load by four to five logs, but the efficacy depends on the MAP concentration, which depends on the prevalence among contributing herds and individuals. Considerable variation of MAP in bulk tank milk (BTM) and individual cow’s milk (IM) is reported, but factors associated with MAP occurrence in milk at farm level have not been described. This study systematically reviewed published studies aiming at estimating the occurrence of MAP in on-farm BTM and IM by metaanalysis. A total of 692 articles were identified through electronic databases and initially screened using title and abstract. The quality of the 61 potentially relevant articles was assessed using full text and 31 articles were eventually included in the meta-analysis. The apparent prevalence (AP) of MAP in BTM and IM on farm were summarized in relation to strata defined by the test used to identify MAP and the infection status of the herds/ animals. There was considerable inconsistency in the reporting, resulting in missing information potentially explaining the dispersion in the estimated AP. The overall AP and 95% confidence intervals based on PCR and culture of MAP were summarized to 0.10 (0.04–0.22) in BTM and 0.20 (0.12–0.32) in IM. Quantifying the MAP load in test-positive milk samples was not possible because very few articles provided quantitative information on individual samples. ß 2012 Elsevier B.V. All rights reserved.
Keywords: Meta-analysis Milk Paratuberculosis Prevalence Systematic review
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . Literature search. . . . . . . . . . . 2.1. Relevance screening . . . . . . . . 2.2. 2.3. Quality assessment. . . . . . . . . Data extraction . . . . . . . . . . . . 2.4. 2.5. Data analysis . . . . . . . . . . . . . Descriptive analysis . 2.5.1. 2.5.2. Stratification . . . . . . Meta-analysis. . . . . . 2.5.3.
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
* Corresponding author. Tel.: +45 35 33 30 35; fax: +45 35 33 30 22. E-mail addresses: [email protected] (H. Okura), [email protected] (N. Toft), [email protected] (S.S. Nielsen). 0378-1135/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.vetmic.2011.12.019
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
254 254 254 254 254 254 255 255 255 255
254
3.
4.
5.
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification and description of relevant literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. 3.2. Detection limits of the test used and possible MAP load in the test-positive milk samples Apparent prevalence of MAP in BTM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Apparent prevalence of MAP in IM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Apparent prevalence of BTM and IM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Systematic review and meta-analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction Mycobacterium avium subsp. paratuberculosis (MAP) causes a chronic infection in ruminants and other animals, and can lead to significant economical losses due to reduced milk yield, premature culling, reduced body condition and eventual death (Kudahl et al., 2007). As the infection progresses, MAP is spread in the body of infected animals and the animals may shed MAP in milk (Whittington and Sergeant, 2001). Presence of MAP in milk for human consumption is of potential concern due to its possible relationship with Crohn’s disease in humans. Pasteurization and processing can be effective for four to five log reduction of the MAP load, but the initial levels of MAP may be too high for complete killing of the bacteria (Grant et al., 2005; Rademaker et al., 2007). A previous review study qualitatively and descriptively summarized information on the presence of MAP in milk without distinguishing between individual milk (IM), bulk tank milk (BTM) on farm, and raw milk at dairy processing plants (Eltholth et al., 2009). To investigate human exposure to MAP via milk, information on the quantity of MAP contamination at each level in the milk processing chain is necessary. Also knowledge about how the contributions of the MAP load at different levels in the chain impact the overall level of MAP is needed. Besides, information on possible load described in specific terms including test accuracy data may provide input to microbiological exposure assessment (Gardner, 2004). A systematic review and meta-analysis allow us to synthesize information from existing studies and also to evaluate the effect of covariates by testing heterogeneities within groups. Therefore, we systematically reviewed published studies aiming at estimating the occurrence of MAP in on-farm BTM and IM by meta-analysis, and at assessing factors associated with reported MAP variation.
2. Materials and methods The review followed a publicly available guide for systematic reviews in agri-food public health (Sargeant et al., 2005).
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
255 255 256 259 259 260 260 260 261 261
EMBASE via Ovid, ISI Web of Science and PubMed. Search keywords used were: M. avium subsp. paratuberculosis, paratuberculosis, Mycobacterium paratuberculosis, and Johne’s each combined with milk. No language restriction was imposed. In addition to the electronic databases, the proceedings of from 6th to 10th International Colloquia on Paratuberculosis from 1999 to 2009 were searched using the same keywords. The literature search included any study testing raw milk collected on farms; regardless of the specified study objectives; in order to maximize the sensitivity of the search. All identified citations with abstracts from each database were uploaded into a reference management software: RefWorks (www.refworks.com). Duplicates were removed by matching the first author, title and publication year. 2.2. Relevance screening Title and abstracts were screened for relevance by one of the authors (HO) using three inclusion criteria: (1) primary research, (2) articles in English, and (3) milk samples collected on farm and tested for presence of MAP (MAP bacteria or MAP DNA). When the abstract was not available and the relevance could not be determined from the title only, the full article was included in the quality assessment. 2.3. Quality assessment A quality assessment form was created to assess further inclusion of the articles. Full papers were obtained for all potentially relevant articles and quality was assessed by 3 reviewers (HO, NT, SSN) working independently. Disagreements among the reviewers were resolved by consensus. The criteria considered essential for inclusion in the review included: description of the milk sampling, description of the test method, reporting of the numbers of samples tested and test-positive. At this step, the articles were mapped to study level, i.e. the combination of milk samples (BTM or IM), animal species (cow or other ruminants), test used (culture, IS900 PCR, or F57 PCR) and infection status (see Section 2.5.2 for details about this stratification). Studies on animals other than cattle (i.e. sheep, goat and buffalo) were excluded.
2.1. Literature search 2.4. Data extraction Seven available electronic databases were searched for articles published between 1990 and November 2010: AGRICOLA, AGRIS, BIOSIS Reviews, CAB Abstracts and
A data extraction form was created by the authors based on the one presented in the guidelines (Sargeant et al., 2005).
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
The form was pre-tested by the 3 reviewers to ensure the consistency of the data extracted from the articles. Data extracted from the relevant studies included general information of the article, population characteristics, study design, measurement of the outcomes, statistical analysis, and results. In the review, a BTM sample referred to a sample collected from the farm bulk tank of a cow herd on a single test-day. Studies including repeated BTM samples from the same farms were excluded if the samples were obtained less than 1 year apart. The MAP presence in these samples was considered to be attributed to the herd demography, which changes considerably in 1 year. Conversely, samples taken more than 1 year apart were considered uncorrelated and thus included in the study. Studies providing only the number of BTM samples without description on the source herds (e.g. the number of the herds from which the BTM samples were collected) were excluded. An individual milk (IM) sample referred to a sample collected from an individual cow on one test-day. The study unit for IM was an individual cow. Therefore, studies describing milk samples representing each teat of the animals were excluded. Furthermore, in contrast to BTM, studies where the study unit was the test-day sample (i.e. multiple days of sampling per animal) were excluded because samples collected from one individual animal for multiple days are expected to be highly correlated given the chronic nature of the infection. 2.5. Data analysis 2.5.1. Descriptive analysis The descriptive analysis of the extracted data was done by summarizing to tables, with stratification as specified in Section 2.5.2. Some studies included in the descriptive tables were excluded in the following meta-analysis in order to keep equal weight to a set of milk samples. For example, when a set of milk samples was sent to multiple testing laboratories leading to multiple results, all were presented in the descriptive analysis; however, only one randomly selected result was used for the statistical analysis. 2.5.2. Stratification Milk samples collected on farms were characterized by 3 strata: type of milk (IM or BTM), test used for MAP detection (culture or PCR), and infection status of the herds/animals (infected or unknown). Basically two tests were used to detect MAP in milk: detection of MAP bacteria by a culture-based method or MAP DNA by PCR. There are further differences within the tests in terms of practical procedures such as preenrichment and DNA extraction but this information was not included. Detection methods, where the presence of MAP in milk included culture of MAP bacteria on solid or liquid media were grouped into ‘culture’. Detection methods, where the presence represents detection of MAP DNA specific insertion sequences, were grouped into ‘PCR’. There are essential differences within PCR methods in terms of which sequence are aimed at because there are different numbers of sequences present in a MAP cell,
255
ultimately affecting both the sensitivity and the estimated MAP load. The analytical specificity may vary depending on the PCR primers used, but there are no published data on the superiority of any of the insertion sequences used. Therefore, within the PCR group, studies were further grouped by the target insertion sequence. Studies were grouped by the infection status of the herds/animals because MAP shedding in milk depends on the infection status of the herd/animals. Herds/animals were grouped into ‘infected’ when the herds/animals were tested positive by either antibody ELISA or fecal culture prior to or simultaneously with the milk testing. Studies with no information on the infection status of the source herds were described with ‘unknown’ infection status. 2.5.3. Meta-analysis A meta-analysis using a random-effects model was used to estimate apparent prevalence (AP) in milk. Analyses were performed to summarize overall AP by BTM and IM, as well as according to the above stratification. A random-effects model was chosen because the true prevalence was assumed to vary across studies, where the reported AP reflects the true variation, random variation and differences in test characteristics (Dohoo et al., 2009, pp. 739–772). Forest plots were used to show the estimated AP of MAP in milk with corresponding 95% confidence intervals in the individual studies and based on a random-effects model (Viechtbauer, 2010). Raw proportions were logit-transformed to avoid negative confidence intervals. Cochran’s Q statistic and p-value were calculated to assess heterogeneity (Dohoo et al., 2009, pp. 739–772). The Q statistic was defined as the total dispersion in the estimated AP in which the null hypothesis tested was ‘no heterogeneity’. Furthermore, Higgins’ I2 statistic, defined as the ratio of the true heterogeneity to total variance across the observed estimate, was calculated to quantify the heterogeneity (Borenstein, 2009), because heterogeneity and dispersion due to differences in the test used and the infection status were expected. Meta-analysis of the AP of MAP in milk was done using R and the package ‘metafor’ (R Development Core Team, 2010; Viechtbauer, 2010). 3. Results 3.1. Identification and description of relevant literature Initially, 692 articles were identified through the electronic databases. After the first screening, 61 articles passed the eligibility assessment. In the quality assessment, 30 of these were excluded prior to data extraction for reasons given in Fig. 1 (Stephan et al., 2002; Stratmann et al., 2006, 2002; Caldow et al., 2007; Shankar et al., 2010; Singh et al., 2007b,c; Singh and Vihan, 2004; Barrington et al., 2003; Corti and Stephan, 2002; Grant et al., 2002, 2000; Hasonova et al., 2009; Hermon-Taylor et al., 1999; Kaittanis et al., 2007; Konuk et al., 2007; Metzger-Boddien et al., 2006; O’Reilly et al., 2004; Ronald et al., 2009; Smith et al., 2003; Taddei et al., 2008; Tasara and Stephan, 2005; Wells, 2005; DimareliMalli, 2010; Djønne et al., 2003; Giese and Ahrens, 2000; Ikonomopoulos et al., 2009;
256
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
• • •
•
• • • • •
• • •
Fig. 1. Steps and results of the systematic review and meta-analysis for Mycobacterium avium subsp. paratuberculosis in milk (bulk tank milk: BTM and individual milk: IM) at farm level.
Kumar et al., 2008; Muehlherr et al., 2003; Nebbia et al., 2006; Selvam et al., 2009; Seyyedin et al., 2008, 2010; Wilson et al., 2010; Favila-Humara et al., 2010; Dzieciol et al., 2010; Ridge et al., 2010; Herthnek et al., 2008; Pillai and Jayarao, 2002; Slana et al., 2008). Eight studies from 5 articles were excluded from the descriptive tables, and studies from 2 articles were further divided as specified in Fig. 1. As a result, the descriptive tables included 26 BTM (Table 1) and 31 IM studies (Table 2). The meta-analysis included 18 BTM and 27 IM studies, because 8 BTM and 4 IM studies were excluded due to multiple sampling (Herthnek et al., 2008; Pillai and Jayarao, 2002;
Slana et al., 2009b), lack of source herd information (Dzieciol et al., 2010; Stratmann et al., 2002) and testing at multiple labs (Ridge et al., 2010) (Tables 1 and 2, Fig. 1). 3.2. Detection limits of the test used and possible MAP load in the test-positive milk samples Information on the test used to analyze the milk samples (e.g. sensitivity, specificity, use of positive/ negative controls, detection limit) and the estimated MAP load was rarely reported. Only 19 studies from 9 articles reported either/both the detection limit and the
Table 1 Apparent prevalence of Mycobacterium avium subsp. paratuberculosis (MAP) in bulk tank milk on cattle farms. Test used
Number of samples tested
Number of samples positive
Apparent prevalence
Country
Study period
Authors
FC positive Mixed*3 Infected Infected Infected ELISA positive ELISA positive Unknown Unknown Unknown Unknown
20 29 91 91 91 14 52 69 237 220 225
1 2 2 0 0 10 0 5 4 0 2
0.05 0.07 0.02 0.00 0.00 0.71 0.00 0.07 0.02 0.00 0.01
USA USA Australia Australia Australia Central Mexico USA Iran Czech Republic Cyprus Cyprus
NR*1 1999–2000 2006 2006 2006 NR NR 2006–2007 NR 2007 2007–2008
Pillai and Jayarao (2002)*5,17 Jayarao et al. (2004) Ridge et al. (2010)*16 Ridge et al. (2010)*16 Ridge et al. (2010)*16 Favila-Humara et al. (2010) Stabel et al. (2002) Seyyedin et al. (2010) Slana et al. (2009a) Slana et al. (2009b) Botsaris et al. (2010)
FC positive Mixed*3 ELISA positive Infected*2 Infected Mixed*4 Unknown Unknown Unknown Unknown Unknown
20 29 14 52 5 143 16 110 237 220 225
10 8 10 35 4 19 0 12 82 49 50
0.5 0.28 0.71 0.67 0.80 0.13 0.00 0.11 0.35 0.22 0.22
USA USA Central Mexico USA Argentina Denmark Brazil Iran Czech Republic Cyprus Cyprus
NR 1999–2002 NR 2007 2006 NR 2006 NR 2007 2007–2008
Pillai and Jayarao (2002)*7,17 Jayarao et al. (2004) Favila-Humara et al. (2010) Stabel et al. (2002) Slana et al. (2008)*8,17 Herthnek et al. (2008)*6,13,17 Carvalho et al. (2009) Haghkhah et al. (2008) Slana et al. (2009a) Slana et al. (2009b)*6,14 Botsaris et al. (2010)*9
Unknown Unknown Infected
100 220 5
3 14 2
0.03 0.06 0.4
Switzerland Cyprus Argentina
2005 2007 2007
Bosshard et al. (2006)*10 Slana et al. (2009b)*6 Slana et al. (2008)*11,17
7
5
0.71
Germany
NR
Stratmann et al. (2002)*12,15,18
ELISA positive
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Culture-based methods (n = 11) HEYM HEYM BACTEC BACTEC BACTEC HEYM and/or Lowenstein-Jensen HEYM HEYM HEYM HEYM HEYM IS900 PCR-based methods (n = 11) IS900 PCR IS900 PCR IS900 PCR IS900 PCR IS900 PCR IS900 realtime PCR IS900 PCR IS900 nested PCR IS900 quantitative PCR IS900 quantitative PCR IS900 combined phage PCR F57 PCR-based method (n = 3) F57 light cycler-based real-time PCR F57 qPCR F57 realtimePCR Other PCR-based method (n = 1) Phage and peptide-mediated capture ISMAv2 PCR
Infection status of the herd
*1 NR: Not reported. *2: No information on the test used for the classification. *3: 16/29 herds were ELISA and/or FC positive. *4: Environmentally positive and negative. Reported detection limit of the test used: *5: 10–100 cfu/ml milk; *6: 100 organisms/ml; *7: 10 cfu/ml; *8: 5–6 MAP cells per/ml; *9: single units of organisms/ml; *10: 100 cells/ml; *11: 83 MAP cells/ml; *12: 10 MAP/ml. Reported estimated MAP load in the positive samples: *13: rarely exceed 100 organisms/ml; *14: single units to tens of organisms/ml; *15: rarely exceed 100 organisms/ml. *16: The set of samples were sent to 3 laboratories. One randomly selected study was used in the meta-analysis. *17: Excluded from the meta-analysis due to multiple sampling. *18: Excluded from the meta-analysis due to lack of source herd information.
257
258
Table 2 Apparent prevalence of Mycobacterium avium subsp. paratuberculosis (MAP) in individual milk samples. Test used
Culture-based methods (n = 13) HEYM
Infection status of the animals
Number of Number of Apparent Country prevalence samples samples positive tested
Study period
Authors
46
0.35
Canada
NR*1
Gao et al. (2009)*2,7
11 24 211 12 87 26 50 1493 126 14 15 15
6 2 9 7 2 25 42 43 3 4 4 1
0.55 0.08 0.04 0.58 0.02 0.96 0.84 0.03 0.02 0.29 0.27 0.07
Denmark Argentina USA Iran Poland India India USA USA Australia Australia Australia
NR NR NR 2006–2007 NR 2004 NR 1999–2002 NR 2006 2006 2006
Giese and Ahrens (2000) Paolicchi et al. (2003) Pillai and Jayarao (2002)*2,8 Seyyedin et al. (2010) Szteyn et al. (2008) Singh et al. (2007a,b,c) Sharma et al. (2008) Jayarao et al. (2004) Streeter et al. (1995) Ridge et al. (2010)*9 Ridge et al. (2010)*9 Ridge et al. (2010)*9
134
71
0.54
Canada
NR
Gao et al. (2009)*3
11
2
0.18
Denmark NR
Giese and Ahrens (2000)
Nested IS900 PCR 11 IS900 PCR 103 IS900 PCR 24 IS900 PCR 211 IS900 PCR 206 IS900 PCR 56 IS900 PCR 87 IS900 PCR 50 IS900 PCR 1493 IS900 PCR 9 IS900 nested PCR 328 Nested IS900 PCR 98 Nested IS900 PCR 46 IS900 realtime PCR 342 Other PCR-based method (n = 2) F57 realtime PCR Unknown 342 Phage and peptide-mediated capture ISMAv2 PCR Serologically positive for MAP in milk or blood ELISA 9
3 21 0 69 8 10 18 3 201 3 13 17 9 64
0.27 0.20 0.00 0.33 0.04 0.18 0.21 0.06 0.13 0.33 0.04 0.17 0.20 0.19
USA Poland Argentina USA Brazil Iran Poland India USA India USA USA USA Argentina
Buergelt and Williams (2004) Wiszniewska-Laszczych et al. (2009) Paolicchi et al. (2003) Pillai and Jayarao (2002)*4 Carvalho et al. (2009) Soltani et al. (2008) Szteyn et al. (2008) Sharma et al. (2008) Jayarao et al. (2004) Kaur et al. (2010) Pinedo et al. (2008a) Pinedo et al. (2008b) Buergelt and Williams (2004) Slana et al. (2008)*5,10
47 7
0.14 0.78
Argentina 2007 Germany NR
LJ and Dubos broth HEYM. HEYM HEYM HEYM HEYM HEYM HEYM HEYM Radiometric culture Radiometric culture Radiometric culture IS900 PCR-based methods (n = 16) Nested IS900 PCR IS900 PCR
Positive on FC or milk or serum ELISA Positive on FC or milk or serum ELISA Positive on serum ELISA Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Subclinical Unknown
Abbreviations: NR, not reported; FC, fecal culture; LJ, Lo¨wenstein–Jensen medium; HEYM, Herrold’s Egg Yolk Medium. Reported detection limits: *2: 10–100 colony forming units (cfu)/ml milk; *3: 2 cfu/ml; *4: 10 cfu/ml; *5: 5–6 MAP cells/ml; *6: 10 MAP/ml. Reported estimated MAP load in the positive samples: *7: 1–5 cfu/ml; *8: 10–100 cfu/ml. *9: The set of samples were sent to 3 laboratories. One randomly selected study was used in the meta-analysis. *10: Excluded from the meta-analysis due to multiple sampling.
NR NR NR NR NR NR NR NR 1999–2002 NR*1 2004–2006 2003–2004 NR 2007
Slana et al. (2008)*5,10 Stratmann et al. (2002)*6
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
133
Positive on FC or milk or serum ELISA FC positive Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
estimated MAP load (Tables 1 and 2). Hence, the estimated MAP load could not be summarized due to different types of milk samples and tests used. Detection limit by culture was reported in one article to be 10–100 cfu/ml milk (Pillai and Jayarao, 2002). Detection limits of IS900, F57 and ISMAv2 PCR were reported in 7 articles ranging from single units to 100 MAP cells/ml of milk (Bosshard et al., 2006; Botsaris et al., 2010; Gao et al., 2009; Herthnek et al., 2008; Pillai and Jayarao, 2002; Slana et al., 2008; Stratmann et al., 2002). Estimated MAP load in test-positive samples was reported in 5 articles, 3 for BTM and 2 for IM. The 3 articles reported estimated MAP load in BTM samples by either IS900 or F57 PCR to be less than 100 MAP organisms/ml of milk (Bosshard et al., 2006; Herthnek et al., 2008; Slana et al., 2009b), while no estimated MAP load by culture was reported. Estimated MAP load in IM samples was reported to be 10–100 cfu/ml milk by culture (Pillai and Jayarao, 2002) and 1–5 cfu/ml milk by IS900 PCR (Gao et al., 2009). 3.3. Apparent prevalence of MAP in BTM The AP among the BTM samples ranged between 0 and 71%. Most reported AP based on culture were below 0.1; however, one study had very high AP of 0.71 (95% confidence interval (C.I.) 0.48–0.95). The AP from PCR methods varied considerably among the studies. The AP from F57 PCR was significantly lower than that of IS900 PCR (Fig. 2). For 18 cow BTM studies, summarized AP and 95% C.I. were 0.10 (0.01–0.13). Results of the subgroup analyses for culture, IS900 PCR and F57 PCR were 0.03 (0.01–0.13),
259
0.30 (0.16–0.49), and 0.05 (0.03–0.1), respectively. The results of further stratification based on the infection status of the herds within the culture and IS900 PCR are shown in Table 3. For the culture group, there was one study in the infected stratum (positive by ELISA or fecal culture) group with very a high prevalence, which resulted in a higher AP with much wider confidence interval. For the IS900 PCR group, the AP for the infected herd group was significantly higher than the unknown herds. There were considerable heterogeneities for the overall meta-analysis and the following sub-group analyses. However, reduction in the I2 statistics was observed as the studies were grouped into the different strata (Table 3). 3.4. Apparent prevalence of MAP in IM Thirty-one studies were included in the descriptive table (Table 2). Many studies reported infection status of the herd from which the individual cows were sampled, while 4 out of 31 studies reported infection status of the animals from which the sample were collected. The AP of IM samples ranged between 0 and 96%. Overall, for 27 IM studies, the summarized AP of MAP from the meta-analysis was 0.20 (0.12–0.32) (Fig. 3, Table 3). Results of the subgroup analysis for the estimated AP based on the culture and IS900 PCR were 0.23 (0.07–0.54) and 0.16 (0.11–0.25), respectively (Fig. 3, Table 3). Stratification by infection status of the cows showed higher AP for ELISA or FC positive cows than for unknown infection status cows.
Fig. 2. Apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis in bulk tank milk on cattle farms.
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
260
Table 3 Summary of the meta-analysis of apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis (MAP) in bulk tank milk (BTM) and individual milk (IM), stratified by diagnostic test and infection status. Stratification BTM Overall Culture Known infected Unknown status IS900 PCR Known infected Unknown F57 PCR IM Overall Culture Known infected Unknown status IS900 PCR Known infected Unknown status
Estimated AP
95% C.I.
I2
Q (p-value)
18 8 3 5 8 2 6 2
0.10 0.03 0.06 0.01 0.30 0.68 0.22 0.05
0.04, 0.01, 0.00, 0.00, 0.16, 0.56, 0.15, 0.03,
0.22 0.13 0.77 0.08 0.49 0.78 0.31 0.10
97.11 89.12 89.86 70.69 95.12 0.00 83.53 32.25
<0.0001 <0.0001 <0.0001 0.01 <0.0001 0.01 <0.0001 0.22
27a 11 2 9 15 3 12
0.20 0.23 0.39 0.20 0.16 0.36 0.14
0.12, 0.07, 0.24, 0.05, 0.11, 0.17, 0.09,
0.32 0.54 0.57 0.56 0.25 0.61 0.21
97.15 97.39 40.47 97.01 93.52 65.54 91.99
<0.0001 <0.0001 0.20 <0.0001 <0.0001 <0.0001 <0.0001
Number of studies
a One study tested the milk samples using phage and peptide-mediated capture ISMAv2 PCR was included in the overall AP estimate but not in the subgroup analysis.
There were considerable heterogeneities for the overall meta-analysis and the following sub-group analyses. Reduction in the I2 statistics as the studies were grouped by the characteristics was only observed in the positive cows and considerable heterogeneity remained in the unknown infection status cow group (Table 3). 4. Discussion
infection status tended to be lower than the AP among known infected herds/animals which was not surprising. It should not be interpreted as if BTMs and IMs without information on infection status have lower prevalence of MAP, but rather that the information on the infection status (test history) could be used to reduce MAP positivity in milk. Summarized AP for BTM was lower than that of IMs which indicated a dilution effect on BTM. It was not possible to quantify the possible MAP load in milk on farm.
4.1. Apparent prevalence of BTM and IM 4.2. Systematic review and meta-analysis This systematic review and meta-analysis estimated the probability of detecting MAP in milk collected on farms. For BTM, AP from culture was lower than AP from IS900 PCR, while in IM AP from culture was higher than AP from IS900 PCR. There are two likely explanations for the different AP in BTM and IM by culture and IS900 PCR: (1) sensitivity of PCR is higher than culture due to a lower detection limit of PCR; or (2) PCR also detects dead MAP. These two circumstances would have different implications at herd and animal level. An individual excreting MAP in milk would probably excrete a relatively high number of viable bacteria, and the effect of dead MAP would be minor. Consequently, the diagnostic test would not be greatly affected by low analytical sensitivity. Conversely, the concentration in BTM is highly affected by the within-herd prevalence of MAP shedders. The high AP for IS900 PCR may reflect higher analytical sensitivity compared to culture. Quantification of the MAP load in the samples would be needed to resolve this issue. In the sub-group analysis for the infection status, summary AP for both BTM and IM were significantly higher for the infected herds/animals than unknown herds/animals (Table 3). It should be noted that there tends to be larger number of samples within the study for unknown herds/animals and caution is needed when interpreting the results from a small number of studies. ‘Unknown’ infection status can include infected herds/animals. AP among herds/animals of unknown
Our search strategy was not limited to prevalence studies. Due to inclusion of non-prevalence studies, the review had to compromise on the quality of the study designs. Therefore many included studies did not adequately describe the choice of sampling scheme, which could have introduced reporting bias. Even taking the reporting bias into consideration, the overall quality of reporting was poor. Many studies did not report information of the test used, e.g. sensitivity and specificity, and even animal species was not clear in two articles (Dzieciol et al., 2010; Stratmann et al., 2002). Meta-analysis in observational studies is less common than randomized control trials, partly because standardized methods of reporting observational studies are relatively new (von Elm et al., 2007; Gardner et al., 2011). Still, there are studies with good quality of reporting (Gao et al., 2009; Herthnek et al., 2008). Heterogeneities in a meta-analysis can occur due to differences in many factors such as country, breed, on-farm management, test used, and definition of the outcomes. These diversities are referred to as ‘clinical’ or ‘methodological’ heterogeneities and may or may not be responsible for the summary estimate. Clinical heterogeneity is a variation explained by the biological context, while statistical heterogeneity exists when the AP truly differs across the studies (Higgins and Green, 2011; Higgins and Thompson, 2002). A heterogeneity indicator such as
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
261
Fig. 3. Apparent prevalence (AP) of Mycobacterium avium subsp. paratuberculosis in milk from individual cows.
Cochran’s Q statistic does not distinguish between different sources of heterogeneity. Hence, it is often difficult to draw conclusions from a meta-analysis when clinical heterogeneity is expected in the summary estimate. A sub-group analysis is one way to investigate the heterogeneity when clinical differences are expected (Barrington et al., 2003; Dohoo et al., 2009, pp. 739–772). If the milk samples from the same infection statuses of animals/herds are tested by the same test the resulting AP should not vary, although there is a variation due to laboratories. Sub-group analysis by diagnostic test and infection status indicated that because the heterogeneity became smaller these subgroups could partly explain the heterogeneity among the studies. Thus, our summary AP and the similarity and differences among the group and sub-groups could reflect the real situation in the probability of detecting MAP in milk on farm.
5. Conclusion The reported AP and 95% C.I. based on PCR and culture of MAP were summarized to 0.10 (0.04–0.2) in BTM and 0.20 (0.12–0.32) in IM. MAP prevalence in milk from antibody ELISA or fecal culture positive herds/animals seemed to be higher than herds/animals with unknown infection status. Considerable variation was observed among the studies, and the MAP load could not be estimated. The quality of the reporting varied considerably resulting in missing information potentially explaining the dispersion in the estimated AP.
References Barrington, G.M., Gay, J.M., Eriks, I.S., Davis, W.C., Evermann, J.F., Emerson, C., O‘Rourke, J.L., Hamilton, M.J., Bradway, D.S., 2003. Temporal patterns of diagnostic results in serial samples from cattle with advanced paratuberculosis infections. J. Vet. Diagn. Invest. 15, 195–200. Borenstein, M., 2009. Introduction to Meta-analysis. Wiley-Blackwell, Chichester, United Kingdom. Bosshard, C., Stephan, R., Tasara, T., 2006. Application of an F57 sequencebased real-time PCR assay for Mycobacterium paratuberculosis detection in bulk tank raw milk and slaughtered healthy dairy cows. J. Food Prot. 69, 1662–1667. Botsaris, G., Slana, I., Liapi, M., Dodd, C., Economides, C., Rees, C., Pavlik, I., 2010. Rapid detection methods for viable Mycobacterium avium subspecies paratuberculosis in milk and cheese. Int. J. Food Microbiol. 141 (Suppl. 1), S87–S90. Buergelt, C.D., Williams, J.E., 2004. Nested PCR on blood and milk for the detection of Mycobacterium avium subsp paratuberculosis DNA in clinical and subclinical bovine paratuberculosis. Aust. Vet. J. 82, 497–503. Caldow, G., Strain, S.A.J., Chapman, Z., Kemp, R., Cook, A.J., 2007. A survey to estimate the herd level prevalence of paratuberculosis in the dairy herd of the United Kingdom. Cattle Pract. 15, 169–171. Carvalho, I.A., Silva Jr., A., Campos, V.E., Moreira, M.A., 2009. Short communication: detection of Mycobacterium avium subspecies paratuberculosis by polymerase chain reaction in bovine milk in Brazil. J. Dairy Sci. 92, 5408–5410. Corti, S., Stephan, R., 2002. Detection of Mycobacterium avium subspecies paratuberculosis specific IS900 insertion sequences in bulk-tank milk samples obtained from different regions throughout Switzerland. BMC Microbiol. 2, 15. DimareliMalli, Z., 2010. Detection of Mycobacterium avium subsp. paratuberculosis in milk from clinically affected sheep and goats. Int. J. Appl. Res. Vet. Med. 8, 44–50. Djønne, B., Jensen, M.R., Grant, I.R., Holstad, G., 2003. Detection by immunomagnetic PCR of Mycobacterium avium subsp. paratuberculosis in milk from dairy goats in Norway. Vet. Microbiol. 92, 135–143. Dohoo, I., Martin, W., Stryhn, H., 2009. Veterinary Epidemiologic Research. VER Inc., Charlottetown, Canada, pp. 739–772.
262
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263
Dzieciol, M., Volgger, P., Khol, J., Baumgartner, W., Wagner, M., Hein, I., 2010. A novel real-time PCR assay for specific detection and quantification of Mycobacterium avium subsp, paratuberculosis in milk with the inherent possibility of differentiation between viable and dead cells. BMC Res. Notes 3, 251. Eltholth, M.M., Marsh, V.R., Van Winden, S., Guitian, F.J., 2009. contamination of food products with Mycobacterium avium paratuberculosis: a systematic review. J. Appl. Microbiol. 107, 1061–1071. Favila-Humara, L.C., Chavez-Gris, G.G., Carrillo-Casas, E.M., HernandezCastro, R., 2010. Mycobacterium avium subsp. paratuberculosis detection in individual and bulk tank milk samples from bovine herds and caprine flocks. Foodborne Pathog. Dis. 7, 351–355. Gao, A., Odumeru, J., Raymond, M., Hendrick, S., Duffield, T., Mutharia, L., 2009. Comparison of milk culture, direct and nested polymerase chain reaction (PCR) with fecal culture based on samples from dairy herds infected with Mycobacterium avium subsp. paratuberculosis. Can. J. Vet. Res. 73, 58–64. Gardner, I.A., 2004. An epidemiologic critique of current microbial risk assessment practices: the importance of prevalence and test accuracy data. J. Food Prot. 67, 2000–2007. Gardner, I.A., Nielsen, S.S., Whittington, R.J., Collins, M.T., Bakker, D., Harris, B., Sreevatsan, S., Lombard, J.E., Sweeney, R., Smith, D.R., Gavalchin, J., Eda, S., 2011. Consensus-based reporting standards for diagnostic test accuracy studies for paratuberculosis in ruminants. Prev. Vet. Med. 101, 18–34. Giese, S.B., Ahrens, P., 2000. Detection of Mycobacterium avium subsp. paratuberculosis in milk from clinically affected cows by PCR and culture. Vet. Microbiol. 77, 291–297. Grant, I.R., Hitchings, E.I., McCartney, A., Ferguson, F., Rowe, M.T., 2002. Effect of commercial-scale high-temperature, short-time pasteurization on the viability of Mycobacterium paratuberculosis in naturally infected cows’ milk. Appl. Environ. Microbiol. 68, 602–607. Grant, I.R., Pope, C.M., O‘Riordan, L.M., Ball, H.J., Rowe, M.T., 2000. Improved detection of Mycobacterium avium subsp. paratuberculosis in milk by immunomagnetic PCR. Vet. Microbiol. 77, 369–378. Grant, I.R., Williams, A.G., Rowe, M.T., Muir, D.D., 2005. Efficacy of various pasteurization time-temperature conditions in combination with homogenization on inactivation of Mycobacterium avium subsp. paratuberculosis in milk. Appl. Environ. Microbiol. 71, 2853–2861. Haghkhah, M., Ansari-Lari, M., Novin-Baheran, A.M., Bahramy, A., 2008. Herd-level prevalence of Mycobacterium avium subspecies paratuberculosis by bulk-tank milk PCR in Fars province (Southern Iran) Dairy Herds. Prev. Vet. Med. 86, 8–13. Hasonova, L., Trcka, I., Babak, V., Rozsypalova, Z., Pribylova, R., Pavlik, I., 2009. Distribution of Mycobacterium avium subsp. paratuberculosis in tissues of naturally infected cattle as affected by age. Vet. Med. 54, 257–269. Hermon-Taylor, J., Cheng, J., Bull, T., 1999. Mycobacterium avium subsp. paratuberculosis in milk: an animal and human pathogen (meeting abstract). J. Microbiol. Methods 38, 217. Herthnek, D., Nielsen, S.S., Lindberg, A., Bolske, G., 2008. A robust method for bacterial lysis and DNA purification to be used with real-time PCR for detection of Mycobacterium avium subsp. paratuberculosis in milk. J. Microbiol. Methods 75, 335–340. Higgins, J.P., Green, S., 2011. Cochrane handbook for systematic reviews of interventions version 5.1.0 [updated March 2011]. The Cochrane Collaboration. Available at: http://www.cochrane.org/training/ cochrane-handbook. Higgins, J.P., Thompson, S.G., 2002. Quantifying heterogeneity in a metaanalysis. Stat. Med. 21, 1539–1558. Ikonomopoulos, J., Liandris, E., Andreadou, M., Gazouli, M., 2009. Associations between abortion-records in goats and test-positivity to Mycobacterium avium subsp. paratuberculosis. Open Vet. Sci. J. 3, 1–5. Jayarao, B.M., Pillai, S.R., Wolfgang, D.R., Griswold, D.R., Rossiter, C.A., Tewari, D., Burns, C.M., Hutchinson, L.J., 2004. Evaluation of IS900-PCR assay for detection of Mycobacterium avium subspecies paratuberculosis infection in cattle using quarter milk and bulk tank milk samples. Foodborne Pathog. Dis. 1, 17–26. Kaittanis, C., Naser, S.A., Perez, J.M., 2007. One-step, nanoparticlemediated bacterial detection with magnetic relaxation. Nano Lett. 7, 380–383. Kaur, P., Filia, G., Singh, S.V., Patil, P.K., Sandhu, K.S., 2010. Molecular detection and typing of Mycobacterium avium subspecies paratuberculosis from milk samples of dairy animals. Trop. Anim. Health Prod. 42, 1031–1035. Konuk, M., Korcan, E., Dulgerbaki, S., Altindis, M., 2007. Isolation and identification of Mycobacteria from raw milk samples in Afyonkarahisar District of Turkey. Int. J. Food Microbiol. 115, 343–347. Kudahl, A.B., Sorensen, J.T., Nielsen, S.S., Ostergaard, S., 2007. Simulated economic effects of improving the sensitivity of a diagnostic test in paratuberculosis control. Prev. Vet. Med. 78, 118–129.
Kumar, S., Singh, S.V., Sevilla, I., Singh, A.V., Whittington, R.J., Juste, R.A., Sharma, G., Singh, P.K., Sohal, J.S., 2008. Lacto-prevalence, genotyping of Mycobacterium avium subspecies paratuberculosis and evaluation of three diagnostic tests in milk of naturally infected goatherds. Small Ruminant Res. 74, 37–44. Metzger-Boddien, C., Khaschabi, D., Schonbauer, M., Boddien, S., Schlederer, T., Kehle, J., 2006. Automated high-throughput immunomagnetic separation-PCR for detection of Mycobacterium avium subsp. paratuberculosis in bovine milk. Int. J. Food Microbiol. 110, 201–208. Muehlherr, J.E., Zweifel, C., Corti, S., Blanco, J.E., Stephan, R., 2003. Microbiological quality of raw goat’s and ewe’s bulk-tank milk in Switzerland. J. Dairy Sci. 86, 3849–3856. Nebbia, P., Robino, P., Zoppi, S., De Meneghi, D., 2006. Detection and excretion pattern of Mycobacterium avium subspecies paratuberculosis in milk of asymptomatic sheep and goats by nested-PCR. Small Ruminant Res. 66, 116–120. O’Reilly, C.E., O’Connor, L., Anderson, W., Harvey, P., Grant, I.R., Donaghy, J., Rowe, M., O’Mahony, P., 2004. Surveillance of bulk raw and commercially pasteurized cows’ milk from approved Irish liquid-milk pasteurization plants to determine the incidence of Mycobacterium paratuberculosis. Appl. Environ. Microbiol. 70, 5138–5144. Paolicchi, F.A., Zumarraga, M.J., Gioffre, A., Zamorano, P., Morsella, C., Verna, A., Cataldi, A., Alito, A., Romano, M., 2003. Application of different methods for the diagnosis of paratuberculosis in a dairy cattle herd in Argentina. J. Vet. Med. B: Infect. Dis. Vet. Public Health 50, 20–26. Pillai, S.R., Jayarao, B.M., 2002. Application of IS900 PCR for detection of Mycobacterium avium subsp. paratuberculosis directly from raw milk. J. Dairy Sci. 85, 1052–1057. Pinedo, P.J., Rae, D.O., Williams, J.E., Donovan, G.A., Melendez, P., Buergelt, C.D., 2008a. Association among results of serum ELISA, faecal culture and nested PCR on milk, blood and faeces for the detection of paratuberculosis in dairy cows. Transbound Emerg. Dis. 55, 125–133. Pinedo, P.J., Williams, J.E., Monif, G.R.G., Rae, D.O., Buergelt, C.D., 2008b. Mycobacterium paratuberculosis shedding into milk: association of ELISA seroreactivity with DNA detection in milk. Int. J. Appl. Res. Vet. Med. 6, 137–144. R Development Core Team, 2010. R: A Language and Environment for Statistical Computing. Rademaker, J.L., Vissers, M.M., Te Giffel, M.C., 2007. Effective heat inactivation of Mycobacterium avium subsp. paratuberculosis in raw milk contaminated with naturally infected feces. Appl. Environ. Microbiol. 73, 4185–4190. Ridge, S.E., Andreata, S., Jones, K., Cantlon, K., Francis, B., Florisson, N., Gwozdz, J., 2010. Inter-laboratory comparison of radiometric culture for Mycobacterium avium subsp. paratuberculosis using raw milk from known infected herds and individual dairy cattle in Victoria. Aust. Vet. J. 88, 249–254. Ronald, B.S.M., Narayanan, R., Krishnakumar, N., Baegan, S., Manohar, B.M., 2009. Detection of Mycobacterium avium subsp. paratuberculosis excretion in goat milk. Asian J. Microbiol. Biotechnol. Environ. Sci. 11, 457–460. Sargeant, J.M., Amezcua, M.D.R., Rajic, A., Waddell, L., 2005. A Guide to Conducting Systematic Reviews in Agri-Food Public Health. Food Safety Research and Response Network. Selvam, A., Vijayarani, K., Kumanan, K., 2009. Detection of Mycobacterium avium subsp paratuberculosis in milk of asymptomatic sheep by polymerase chain reaction. Indian J. Anim. Sci. 79, 784–785. Seyyedin, M., Salehi, T.Z., Tadjbakhsh, H., Najafi, M.F., Rabbani, M., 2008. Isolation and identification of Mycobacterium avium subsp. paratuberculosis from milk, manure and fecal samples of holstein-friesian dairy cattle. Pakistan J. Biol. Sci. 11, 2639–2645. Seyyedin, M., Zahraei, T., Najafi, M.F., 2010. Comparison of isolation frequency of Mycobacterium avium subspecies paratuberculosis from different types of samples. Pakistan Vet. J. 30, 143–149. Shankar, H., Singh, S.V., Singh, P.K., Singh, A.V., Sohal, J.S., Greenstein, R.J., 2010. Presence, characterization, and genotype profiles of Mycobacterium avium subspecies paratuberculosis from unpasteurized individual and pooled milk, commercial pasteurized milk, and milk products in India by culture, PCR, and PCR-REA methods. Int. J. Infect. Dis. 14, e121–e126. Sharma, G., Singh, S.V., Sevilla, I., Singh, A.V., Whittington, R.J., Juste, R.A., Kumar, S., Gupta, V.K., Singh, P.K., Sohal, J.S., Vihan, V.S., 2008. Evaluation of indigenous milk ELISA with m-Culture and m-PCR for the diagnosis of bovine Johne’s disease (BJD) in lactating Indian dairy cattle. Res. Vet. Sci. 84, 30–37. Singh, S.V., Vihan, V.S., 2004. Detection of Mycobacterium avium subspecies paratuberculosis in goat milk. Small Ruminant Res. 54, 231–235. Singh, S.V., Singh, A.V., Singh, R., Sandhu, K.S., Singh, P.K., Sohal, J.S., Gupta, V.K., Vihan, V.S., 2007a. Evaluation of highly sensitive
H. Okura et al. / Veterinary Microbiology 157 (2012) 253–263 indigenous milk ELISA Kit with fecal culture, milk culture and fecalPCR for the diagnosis of bovine Johne’s disease (BJD) in India. Comp. Immunol. Microbiol. Infect. Dis. 30, 175–186. Singh, S.V., Singh, A.V., Singh, R., Sandhu, K.S., Singh, P.K., Sohal, J.S., 2007b. Comparison of multiple diagnostic tests for the diagnosis of bovine paratuberculosis in lactating cows. Indian J. Anim. Sci. 77, 1115–1117. Singh, S.V., Singh, P.K., Singh, A.V., Sohal, J.S., Subodh, S., Narayanasamy, K., 2007c. Non-chemical method of DNA recovery and characterization of Mycobacterium avium subspecies paratuberculosis using IS900 PCR. Indian J. Exp. Biol. 45, 812–816. Slana, I., Kralik, P., Kralova, A., Pavlik, I., 2008. On-farm spread of Mycobacterium avium subsp. paratuberculosis in raw milk studied by IS900 and F57 competitive real time quantitative PCR and culture examination. Int. J. Food Microbiol. 128, 250–257. Slana, I., Bartos, M., Roubal, P., Babak, V., Pavlik, I., 2009a. Mycobacterium avium subsp. paratuberculosis and M. a. avium detected by culture, IS900 and IS901 highly sensitive PCR in Bulk tank milk from dairy herds in the Czech Republic between 2002 and 2004. Czech J. Food Sci. 27, 372–378. Slana, I., Liapi, M., Moravkova, M., Kralova, A., Pavlik, I., 2009b. Mycobacterium avium subsp. paratuberculosis in cow bulk tank milk in Cyprus detected by culture and quantitative IS900 and F57 realtime PCR. Prev. Vet. Med. 89, 223–226. Smith, W.L., McGarvey, K.L., Cullor, J.S., 2003. The use of spiral plating and microscopic colony counting for the rapid quantitation of Mycobacterium paratuberculosis. Lett. Appl. Microbiol. 36, 293–296. Soltani, M., Nassiry, M.R., Shahroudi, F.E., Bassam, M.R., 2008. Detection of Mycobacterium paratuberculosis in feces and milk samples from holstein dairy cows by PCR. Biotechnology 7, 582–585. Stabel, J.R., Wells, S.J., Wagner, B.A., 2002. Relationships between fecal culture, ELISA, and bulk tank milk test results for Johne’s disease in US dairy herds. J. Dairy Sci. 85, 525–531. Stephan, R., Buhler, K., Corti, S., 2002. Incidence of Mycobacterium avium subspecies paratuberculosis in bulk-tank milk samples from different regions in Switzerland. Vet. Rec. 150, 214–215. Stratmann, J., Dohmann, K., Heinzmann, J., Gerlach, G.F., 2006. Peptide aMptD-mediated capture PCR for detection of Mycobacterium avium
263
subsp. paratuberculosis in bulk milk samples. Appl. Environ. Microbiol. 72, 5150–5158. Stratmann, J., Strommenger, B., Stevenson, K., Gerlach, G.F., 2002. Development of a peptide-mediated capture PCR for detection of Mycobacterium avium subsp. paratuberculosis in milk. J. Clin. Microbiol. 40, 4244–4250. Streeter, R.N., Hoffsis, G.F., Bech-Nielsen, S., Shulaw, W.P., Rings, D.M., 1995. Isolation of Mycobacterium paratuberculosis from colostrum and milk of subclinically infected cows. Am. J. Vet. Res. 56, 1322–1324. Szteyn, J., Wiszniewska-Laszczych, A., Ruszczynska, A., 2008. Effectiveness of Mycobacterium paratuberculosis isolation from raw milk by means of direct isolation of DNA and classic culture. Pol. J. Vet. Sci. 11, 25–28. Taddei, R., Barbieri, I., Pacciarini, M.L., Fallacara, F., Belletti, G.L., Arrigoni, N., 2008. Mycobacterium porcinum strains isolated from bovine bulk milk: implications for Mycobacterium avium subsp. paratuberculosis detection by PCR and culture. Vet. Microbiol. 130, 338–347. Tasara, T., Stephan, R., 2005. Development of an F57 sequence-based realtime PCR assay for detection of Mycobacterium avium subsp. paratuberculosis in milk. Appl. Environ. Microbiol. 71, 5957–5968. Viechtbauer, W., 2010. Conducting meta-analyses in R with the metafor package. J. Stat. Softw. 36, 1. von Elm, E., Altman, D.G., Egger, M., Pocock, S.J., Gotzsche, P.C., Vandenbroucke, J.P., STROBE Initiative, 2007. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 370, 1453–1457. Wells, S.J., 2005. Is milk or feces more important in transmission of map? In: Proceedings of the North American Veterinary Conference, Large Animal, vol. 19. p. 55. Whittington, R.J., Sergeant, E.S., 2001. Progress Towards understanding the spread, detection and control of Mycobacterium avium subsp paratuberculosis in animal populations. Aust. Vet. J. 79, 267–278. Wilson, D.J., Rood, K., Biswas, P., Byrem, T.M., 2010. Herd-level prevalence of Johne’s disease in Utah and adjacent areas of the intermountain West as detected by a bulk-tank milk surveillance project. J. Dairy Sci. 93, 5792–5797. Wiszniewska-Laszczych, A., Szteyn, J., Smolinska, A., 2009. Analysis of correlations between the occurrence of anti-MAP antibodies in blood serum and the presence of DNA-MAP in milk. Pol. J. Vet. Sci. 12, 379–383.