Abstract
The objective of this study was to systematically collect and appraise the scientific evidence related to risk factors associated with the introduction of Mycobacterium avium spp. paratuberculosis (MAP) into a herd of cattle.
An electronic search was conducted to collect relevant references addressing 2 specific questions: are i) purchasing/introduction of cattle into a herd, and ii) presence of wildlife or domestic animals, risk factors for the introduction of MAP into a herd? The screening was based on titles and abstracts and selected studies were fully analyzed.
Seventeen manuscripts published between 1996 and 2011 were ultimately analyzed. Unit of interest was mainly the herd (n = 17). The specific description of the risk factors studied varied between studies. The principal study design was cross-sectional (n = 15).
The review indicated that purchase/introduction of animals was an important risk factor and that the importance of wildlife or other domestic species as a mechanism for transmission into a cattle herd was not measurable.
Résumé
Revue systématique sur les facteurs de risque associés à l’introduction de Mycobacterium avium spp. paratuberculosis (MAP) dans les troupeaux laitiers. L’objectif était la collecte et l’évaluation systématique des preuves scientifiques liées à des facteurs de risque associés à l’introduction de Mycobacterium avium spp. paratuberculosis (MAP) dans un troupeau.
Une recherche électronique a été réalisée pour trier des publications afin de répondre à 2 questions : Est-ce que i) l’achat/introduction de bovins dans un troupeau, et ii) la présence d’animaux sauvages et domestiques sont des facteurs de risque pour l’introduction de MAP dans un troupeau. La sélection a été effectuée en se basant sur les titres, les résumés, et les études choisies ont été entièrement analysés.
Dix-sept manuscrits publiés entre 1996 et 2011 ont été sélectionnés. L’unité d’intérêt a principalement été le troupeau (n = 17). Les descriptions des facteurs de risque étudiés variaient d’une étude à l’autre. La majorité des études étaient de type transversal (n = 15).
L’achat d’animaux est un facteur de risque bien documenté confirmant l’introduction de la paratuberculose dans le troupeau. Toutefois, malgré la présence de MAP dans les populations domestiques et sauvages à proximité du bétail, l’importance de ce facteur de risque dans la transmission à un troupeau n’était pas mesurable.
(Traduit par les auteurs)
Introduction
Mycobacterium avium ssp. paratuberculosis (MAP) is the causative agent of paratuberculosis, also called Johne’s disease (JD) in ruminants. Paratuberculosis is an enteric, chronic, incurable disease of ruminants that occurs most commonly in dairy cattle (1) and causes significant economic loss (2). An association between Crohn’s disease in humans and MAP has been described, although causality remains to be proven (3,4).
Transmission of MAP in ruminants is primarily through the fecal-oral route (5), although infection can also occur in utero (6) or, via ingestion of contaminated colostrum or milk (7,8). However, prior to controlling for within-herd transmission, it is relevant to explore how the herd becomes infected. Purchase or introduction of cattle into a herd (9–11) is reported as the major risk factor associated with introduction of the disease. The animal introduced may be subclinically infected (11,12), shed the bacteria, and contaminate the environment for several months to years before clinical signs appear.
Contact with feces of infected animals including wildlife, such as rabbits (13,14), is also a potential risk factor for the introduction of MAP into herds. Mycobacterium avium ssp. paratuberculosis has been isolated from various ruminant and non-ruminant wild species (14–18) and in domestic animals such as sheep and goats (4) that may be co-grazing with infected cattle (9). In one study, the strain isolated in wild animals was the same as in cattle (19) showing the potential role of wildlife in the epidemiology and transmission of MAP to livestock. We plan to examine the evidence available supporting the claim that the presence of wildlife or domestic animals on the farm/pasture is a risk factor for transmission of MAP into a herd of cattle.
Due to the long course of the disease and the difficulty of early definitive diagnosis (4), prevention of the introduction of MAP combined with reduction of the risk of transmission through adequate herd management practices seem the most important keys to control of the disease. The objective of this study was to systematically collect and appraise the scientific evidence related to 2 specific risk factors associated with the introduction of MAP into a herd: i) Is introducing cattle a risk factor? and ii) Is the presence of wild or domestic animals (e.g., deer, rabbits, sheep, goats, cats) on the farm/pasture a risk factor?
Materials and methods
The systematic review was conducted based on the guidelines from “Systematic Reviews” (20), “The process of systematic review and its application in agri-food public-health” (21), and “Systematic reviews to support evidence-based medicine” (22).
Search strategy
The first search was conducted in July 2011. Four online databases were selected: CAB from 1973 to 2011, Medline including PubMed from 1946 to 2011, Embase from 1980 to 2011 (all 3 through OVID platform), and Web of Science from 1979 to 2011. These databases were considered to be relevant and have an important impact on worldwide veterinary research. A second search was conducted in November 2013 to update the list of references using exactly the same search strategy in the 4 databases. A limit was added (year of publication) in order to retrieve only the manuscripts published from 2011 until November 2013.
The search addressed 2 specific questions related to the risk of a dairy herd becoming infected with MAP, and a specific search strategy was designed for each question screening through title, abstract, and keywords.
-
Is purchasing or introducing cattle a risk factor?
Mycobacterium avium subsp. paratuberculosis OR Paratuberculosis OR Johne’s disease AND Cattle OR Bovi* OR Cow* OR Dairy OR Ruminant* OR Calf OR Calves OR Bull* OR Heifer* AND Contaminant* OR Transmi* OR Epidemiolog* AND Risk factor* OR Hazard* AND Purchas* OR Buy* OR Acquir* OR Introduc* OR Rent* OR Borrow*.
-
Is the presence of wildlife or domestic animals (e.g., deer, rabbits, sheep, goats, cats) on the farm/pasture a risk factor?
Mycobacterium avium subsp. paratuberculosis OR Paratuberculosis OR Johne’s disease AND Cattle OR Bovi* OR Cow* OR Dairy OR Ruminant* OR Calf OR Calves OR Bull* OR Heifer* AND Contaminant* OR Transmi* OR Introduc* OR Epidemiolog* AND Grass* OR Pasture* OR Ensilage* OR Silage* OR Forag* OR Hay OR Grazing OR Water AND Wildlife OR Sheep* OR Goat* OR Deer* OR Rabbit* OR Hare* OR Cat OR Cats OR Dog OR Dogs OR Bird* OR Mammal* OR Domestic* animal* OR Cross-species OR Wildlife reservoir OR Inter species AND Faec* OR Feces OR Fecal OR Dropping* OR Dung OR Urine.
Identification of relevant studies
Rules for including studies in the analysis were: papers published in peer-reviewed journals; studies that answer the specific question with scientific elements or describe any relation with the risk factor studied and quantify or measure the risk; manuscripts in English, French, or Spanish. A first screening was done by one author (S.R.) based on the titles; selected manuscripts were submitted to a second selection by reading the abstracts (S.R.). The criteria for excluding studies from the analysis were: i) studies unrelated to our questions; ii) studies on economic impact, genetics, diagnostic tests, prediction models, human medicine, vaccine, embryo transfer, and nutrition; and iii) studies of MAP infection in species other than cattle. The studies that were included were fully and independently reviewed by 3 authors (S.R., G.F., J.A.), to assess their relevance according to the information (evidence) provided to answer the questions.
Data extraction
General information was collected from each article: author, year of publication, country where the study was performed, study design, unit of interest and number of subjects (animals or herds), definition of a case (positive animal or positive herd) and the diagnostic test used. The sampling strategy, risk factor studied, statistical analysis used, results and conclusions regarding significant or non-significant association related to the questions were also recorded.
Study appraisal
Manuscripts that met the inclusion criteria were considered relevant in the final selection.
Based on the criteria used by Doré et al (5) derived from a qualitative checklist used by Sanderson et al (23), we applied the following criteria for the study appraisal. To assess the internal validity, 3 factors were considered: the study design, the diagnostic test reliability, and the case definition. Studies were classified as having a high internal validity if the study was longitudinal or experimental, the diagnostic test detected MAP by culture, PCR, or MAP antigen, and the case definition was clearly stated. Studies were classified as having a moderate internal validity if it was a cross-sectional or case-control study with a diagnostic test detecting MAP antigen, OR a longitudinal study with a diagnostic test detecting a humoral response to MAP. All other studies were classified as having a low internal validity.
To assess external validity, the sampling strategy and the sample size (40 herds/cows as an arbitrary cut point) (5) were studied. Studies with < 40 herds or cows enrolled in a voluntary control program were classified as low, studies with ≥ 40 enrolled herds/cows in a voluntary control program or fewer herds/cows for which the sampling was done randomly were classified as moderate, and studies having > 40 herds/cows and a random sampling were classified as high (5).
Results
A total of 17 manuscripts were kept for data extraction and analysis. The articles were published in 11 journals and the studies were carried out in 9 countries, mainly in Europe (n= 9), and the United States (n = 6). The unit of interest was the herd in all studies (n = 17) but some studies also examined individual level data (n = 5). Enzyme-linked immunosorbent assay (ELISA) was the most frequently used diagnostic test (n = 9). There were 7 types of ELISA. The second most frequently used test was bacteriological culture (n = 7). Some studies (n = 5) used > 1 test to determine the infection status. The specific description of the risk factor studied varied from study to study. The principal study design was cross-sectional (n = 15). The data extracted are listed in Table 1. Qualitative study appraisal for internal and external validity is presented in Table 2. Studies were classified as low for internal validity (n = 9), or moderate (n = 8); the majority were classified as moderate (n = 10) for external validity.
Table 1.
General data extracted from the articles included in the analysis
| Question | Author (reference) | Year | Country | Unit of interest Number of subjects | Sample strategy | Case definition | Diagnostic test |
|---|---|---|---|---|---|---|---|
| 1 | Goodger (24) | 1996 | United States | 26 herds | Convenience, positive herd history | Positive herd: 1 clinical case/year or 2 or more IFC (+) in the past year | ELISA serum IDEXXa |
| 1 | Obasanjo (25) | 1997 | Unites States | 33 herds | Convenience from control program | Infected herd: 1 or more MAP (+) FC results | Fecal culture in HEYMb |
| 1–2 | Çetinkaya (33) | 1997 | England | 3772 herds | Random, stratified 3 regions | Cases reported in 1993/1994 | N/A |
| 2 | Greig (35) | 1999 | Scotland | 22 farms (210 rabbits) | Convenience, trapped animals | Farm history with or without PTB. Rabbits MAP culture (+) or lesions at necropsy |
Histopathology, fecal culturec, PCR |
| 1 | Wells (12) | 2000 | United States | 1004 herds (32 622 cows) | Stratified random sample from control program | Positive herds: > 2 cows (+) or 1 cow (+) and ≥ 5% of culled cows with JD signs in the previous year | ELISA serum IDEXXa |
| 1–2 | Daniels (29) | 2002 | Scotland | 86 herds | Convenience but effort to unbiased | History of clinical cases: herd with and without the disease | N/A |
| 1–2 | Muskens (30) | 2003 | The Netherlands | 370 herds | Random | Positive herd: 1 ELISA test (+) | ELISA serum IDEXXd |
| 1 | Hirst (28) | 2004 | United States | 15 herds (10 280 cows) | Convenience, volunteers | Infected herd: ≥ 1 individual fecal culture (+) or ≥ 1 culled cow with histologic evidence of MAP infection | ELISA IDEXXa, modified fecal culture in HEYM, PCR, histopathology |
| 1 | Orpin (10) | 2005 | United Kingdom | 23 herds | Convenience, target sampling | Clinical history: Heavily infected (≥ 2 clinical cases/5 years); single positive (1 isolated case in 5 years) | ELISA milk/serum Pourquiere |
| 2 | Corn (34) | 2005 | United States | 9 herds [2752 wild species samples (feces, tissues)] | Convenience, trapped animals/intensive sampling | Herd prevalence: consecutive FC and serologic tests | Fecal culture (Bactec), genotyping, PCR-sequencingf, histopathology |
| 2 | Raizman (36) | 2005 | United States | 80 case herds 28 control herds |
Convenience from control program, cows/wildlife fecal samples | Herd, environment, or wildlife positive if at least 1 sample was + to MAP culture | Fecal cultureg ELISA serum IDEXXh |
| 1 | Nielsen (32) | 2007 | Denmark | 97 herds | Convenience | Continuous optical density (OD) milk ELISA | In-house milk ELISAi |
| 1 | Cashman (27) | 2008 | Ireland | 59 herds (2333 cattle serum samples) | Convenience, apparently clinical history | Positive herd = One culture or PCR (+) test of milk sock filter/6 | MFR culture, PCR ELISA serum Pourquiere |
| 1 | Tiwari (31) | 2009 | Canada | 315 herds | Random, stratified 2 stages | ELISA (+) = cow (+) | ELISA serum IDEXXh/BIOCORj |
| 1 | Ansari-Lari (26) | 2009 | Iran | 110 herds | Convenience | Positive herd = PCR (+) on bulk tank milk | PCR IS900 bulk tank milk |
| 1 | Nielsen (7) | 2011 | Denmark | 1081 herds | Convenience from control program | High risk animal: milk ELISA positive at least once within the last 3 tests | ELISA milk ID Screen ID-VETk |
| 1–2 | Barrett (9) | 2011 | Ireland | 86 case herds 125 control herds |
Convenience | Positive herd: 1 or > MAP (+) IFC results Control herd: no MAP(+) |
Fecal culture |
All studies were cross-sectional except those by Raizmen et al (36) and Barrett et al (9), which were case-control. IFC — Individual fecal culture; FC — fecal culture; JD — Johne’s disease; PTB — paratuberculosis; OD — optical density; MAP — Mycobacterium avium ssp. paratuberculosis; HEYM — Herrold’s egg yolk medium; MFR — Milk sock filter residue; PCR — polymerase chain reaction; N/A — not applicable.
Mycobacterium paratuberculosis antibody test kit, IDEXX Laboratories, Westbrook, Maine. USA.
Cornell double incubation method.
Middlebrook 7H11 agar supplemented with Mycobactin J (Allied Monitor, Fayette, Missouri, USA) and Selectatabs (code MS24, MAST Laboratories, Merseyside, UK).
Herdchek Mpt AB, Idexx Skandinavia AB, Sweden.
Paratuberculosis ELISA kit, Institute Pourquier, Montpellier, France.
16S/IS900/IS1311 and L1/L9 integration site PCR.
Reference 48.
IDEXX Herdchek ELISA; IDEXX Laboratories.
Reference 49.
BIOCOR Paracheck ELISA; BIOCOR Animal Health Inc., Omaha, Nebraska, USA.
IDScreen, ID-Vet, Montpellier, France.
Table 2.
Study appraisal results
The first search in all databases concerning purchase/introduction of cattle (Question 1, Figure 1) yielded 536 citations. These were screened based on the title and 425 were considered irrelevant. From the remaining 111 citations, 40 were discarded because they were duplicates between databases. Subsequently, the abstracts of 71 manuscripts were read: 46 were discarded because they did not answer or address the question, and 4 were discarded because they were not in English, French, or Spanish. A total of 21 articles were fully read, and 7 were excluded because they did not measure the impact of the risk, finally 14 (7,9,10,12,24–33) were selected for the analysis.
Figure 1.
Flowchart for selection of articles (search 1).
The second search (2011–2013) yielded 10 new citations. Following the same selection process and using the same inclusion/exclusion criteria, 3 citations were excluded because they were duplicates. Screening the titles, 2 manuscripts were selected and after reading the abstracts, 1 was considered irrelevant. The remaining manuscript was fully read and was excluded from the analysis because it did not measure the impact of the risk.
Of the 14 papers that were finally selected (Figure 1, Table 3), a significant association between the risk of being a MAP-contaminated herd related to introducing cattle was found in 6 articles (7,9,12,28,31,33). The proportion of purchased cattle in the herd (7) (> 15%), the percentage of cows born at other dairies (12) (≥ 25%), and the annual importation rate (28) (≥ 8%), were found to increase the risk of a herd being classified as MAP-infected. These proportions were measured by: the measure of MAP prevalence [odds ratio (OR): 3.8 in a 4-year period, P < 0.001]; herd status by prior diagnosis of JD infection [OR: 2.9, 95% confidence interval (CI): 1.6 to 5.1] and seropositivity by ELISA test (OR: 3.28, 95% CI: 2.15 to 5.03), respectively. A case control study conducted in Ireland by Barrett et al (9) found an increased risk of detection of MAP (19.22 times more likely, 95% CI: 2.87 to ∞) in herds with direct cattle importation from continental Europe. The variable “open heifers purchased during the last 12 months” was associated with having a greater count of MAP-seropositive cows (CR: 2.3, 95% CI: 1.2 to 3.5). However, the variable “bulls purchased during the last 12 months” was associated with fewer MAP-seropositive cows (CR: 0.6, 95% CI: 0.4 to 0.9), being the first report of a negative association for purchasing cattle (31). Çetinkaya et al (33) reported that the purchase of replacement privately (from several sources) was associated with an increased risk of reporting the disease (OR: 4.19, 95% CI: 1.77 to 9.93).
Table 3.
Results of manuscripts that studied the risk factor concerning purchase of cattle (question 1)
| Author (reference) | Risk factor studied/statistical analysis | Results and conclusions |
|---|---|---|
| Orpin (10) | 1. Policy for purchasing replacement stock: high, medium, low risk related to the risk of having JD. Descriptive statistics. |
A herd with a high risk score is more likely to have clinical disease compared to herds with low risk (OR 2.6; 95% CI: 0.4 to 17.5) but it was not significant. |
| Goodger (24) | 1. No purchased cattle in the previous year (replacement source). Multiple linear regression and logistic regression. |
Scores on the replacement source category did not influence the variability of MAP apparent prevalence (results not significantly different). |
| Obasanjo (25) | 1. Clinical disease on purchased animals. Univariate/Multivariate stepwise logistic regression. |
Not significant at multivariate stepwise logistic regression (OR: 1.7; 95% CI: 0.1 to 31.9). |
| Nielsen (7) | 1. Proportion of purchased animals in the herd (No purchased animals; 0% to 15% and > 15%). Univariate/Multivariate analysis — linear mixed model. |
The higher proportion of purchased animals (> 15%) had more influence on changes in prevalence (P < 0.001). |
| Ansari-Lari (26) | 1. Purchase replacement animals (yes or no). Univariate/Multivariate logistic regression. |
Not significant in univariate analysis and not included in multivariate model. |
| Cashman (27) | 1. Purchased animals (and contact with imported or progeny of imported cattle). Univariate/Logistic regression. |
None of the risk factors related to purchase (biosecurity) was significantly associated with the probability of a positive MFR culture or PCR. |
| Barrett (9) | 1. Direct cattle importation. Univariate/Multivariate logistic regression. |
The variable was associated significantly (P < 0.05) with detection of MAP in the univariate analysis and in the multivariate exact logistic regression analysis (OR: 19.22; 95% CI: 2.8 to ∞; P = 0.001). |
| Hirst (28) | 1. Annual importation rate (< 8% versus ≥ 8%) in preceding 5 years. Logistic regression/Pearson correlation coefficients. |
Positive correlation between importation rate and prevalence of seropositive cattle (r = 0.87). Cows in herds with ≥ 8% importation were more likely to be seropositive for MAP (OR: 3.28; 95% CI: 2.15 to 5.03; P = 0.03). |
| Daniels (29) | 1. Number of cattle bought in. 2. Change in number of cattle bought. Univariate/Multivariate stepwise regression. |
The presence of the disease was not related to the absolute numbers of livestock bought in. |
| Muskens (30) | 1. Purchased dairy cattle during the last 3 and 10 years. Univariate/Multivariate analysis. |
No significant differences were observed between the seronegative and seropositive groups. |
| Wells (12) | 1. Percentage of cows born at other dairies (0%; 0.1% to 24%; ≥ 25%). Univariate/Multivariate logistic regression and X 2. |
Herds with ≥ 25% of cows born on other dairies were more likely to have a positive herd status (OR: 2.1; 95% CI: 1.3 to 3.5) and were also associated with prior diagnosis of JD (OR: 2.9; 95% CI: 1.6 to 5.1). |
| Tiwari (31) | 1. Open heifers purchased during the last 12 months. 2. Open bulls purchased during the last 12 months. ZINB multivariate model. |
The variable “open heifers purchased” was positively associated with the number of MAP seropositive cows (CR: 2.3; 95% CI: 1.2 to 3.5). “Bulls purchased” was associated with a lower count of MAP seropositive cows (CR: 0.6; 95% CI: 0.4 to 0.9). |
| Nielsen (32) | 1. Purchased cows in herd 2. More than > 5% purchased cows in herd. Finite mixture model (multilevel Bayesian). |
Purchased cows (OR: 1.27; 95% CI: 0.65 to 2.9) and more than 5% (OR: 1.34; 95% CI: 0.74 to 2.4) were not statistically significant. |
| Çetinkaya (33) | 1. Source of replacement: purchased from markets, from dealers, purchased privately. 2. Replacement rate: < 10%; 11% to 20%; > 20%. Univariate/Multivariate regression analysis. |
Purchasing replacements privately was significantly associated with an increased risk of reporting disease (OR: 4.19; 95% CI: 1.77 to 9.93; P = 0.001). Replacement rate was not significantly associated. |
MAP — Mycobacterium avium ssp. paratuberculosis; JD — Johne’s disease; OR — odds ratio; CR — count ratio; ZINB — Zero-inflated negative binomial model; CI — confidence interval.
All other studies (n = 8) examining the effect of introducing cattle into a herd (10,24–27,29,30,32) did not find a statistically significant association with the risk of MAP infection. Nielsen and Toft (32) identified a tendency for purchasing cattle (OR: 1.27, 95% CI: 0.65 to 2.9), or having more than 5% purchased cattle in herd (OR: 1.34, 95% CI: 0.74 to 2.4), to be associated with MAP prevalence, but neither achieved statistical significance.
Orpin et al (10) categorized the risk associated with purchasing replacement stock according to the number and frequency of purchases in the last 13 years in high, medium, and low levels in relation to the risk of presence of disease. They found that a herd with a high risk score was approximately 3 times more likely to have clinical disease (OR: 2.6) compared to herds with lower risk; however, it was not statistically significant (95% CI: 0.4 to 17.5).
The first search in all databases for the question related to risk of presence of wildlife or other domestic animals (Question 2, Figure 1) yielded 967 citations. These citations were screened based on the title and 868 were considered irrelevant. From the remaining 99 citations, 55 were discarded because they were duplicates between the 4 databases. The abstracts were read and 24 were discarded because they did not answer or address the question or because they were not related to cattle. Twenty articles were fully read and 13 were excluded because they did not examine the impact of the risk. Finally 7 (9,29,30,33–36) were selected for the analysis (Table 4), 4 of which were duplicates from question 1 (Figure 1). The second search (2011–2013) yielded 15 new citations, 6 of which were discarded because they were duplicates. From the remaining 9 manuscripts, 8 were excluded because the titles did not match the inclusion criteria and 1 manuscript was not kept for analysis because the study was conducted in farmed deer.
Table 4.
Results and conclusions of manuscripts that studied the risk factor concerning wildlife and other domestic animal exposures (question 2)
| Author (reference) | Risk factor studied/statistical analysis | Results and conclusions |
|---|---|---|
| Daniels (29) | 1. Presence of small ruminants, other domestic animals and wildlife in the farm. 2. Wildlife access to stored feed. Univariate analysis/Multivariate stepwise regression. |
Univariate: 1. Total number of sheep and cattle, number of rabbits present, number of crows present; 2. Access of wildlife to stored farm mix, were all associated with the probability of a farm reporting JD. Final model: 1. Number of rabbits (few vs many, 0.77 MP; P = 0.018); 2. Access of wildlife to stored farm mix (0.899 MP; P = 0.028) were all associated with the probability of a farm reporting JD. |
| Muskens (30) | 1. Presence sheep/goats. 2. Grazed sheep other farmer. Univariate/multivariate analysis. |
The variables were not significantly associated. |
| Corn (34) | 1. Interspecies transmission. Descriptive, X 2. |
There was no statistical correlation between the test-prevalence in livestock vs wildlife on the farms (r = 0.31). There was no difference in prevalence of MAP infection in wildlife among all farms (X2 = 14.43, P > 0.10). |
| Greig (35) | 1. Association MAP in rabbits versus farm with MAP. Linear modeling LOGIT. |
Statistically significant relationship between a past or current problem of paratuberculosis in cattle and in the wild rabbit population (P = 0.013). |
| Barrett (9) | 1. Co-grazing of heifers with sheep and/or goats. Univariate/Multivariate logistic regression. |
The variable was significantly associated (P < 0.05) with detection of MAP in the univariate analysis. |
| Çetinkaya (33) | 1. Presence of other species on the farm: sheep, goats, and farmed deer. Univariate/Multivariate regression analysis. |
Presence of farmed deer was found to be highly associated with clinical disease only for 1 year of the 2 years studied (OR: 209.3; P = 0.001 and 95% CI: 13.47 to 3253). |
| Raizman (36) | 1. Relation between MAP in wildlife and infected herds. 2. Use of pasture by cattle and the risk of contact between cattle manure and deer/rabbit. Descriptive analysis, X2 fisher’s test, Student’s t-test. |
1. There was no significant association between MAP in wildlife and infected herds despite the significant association found between the use of pasture by cattle and the risk of contact between cattle manure and both deer and rabbit (OR: 5.4, 95% CI: 2.1 to 14.2 and OR: 3.6, 95% CI: 1.3 to 9.5, respectively). |
MP — mean probability; MAP — Mycobacterium avium ssp. paratuberculosis; JD — Johne’s disease; CI — confidence interval.
A significant positive association between the presence of wildlife and domestic animals at the farm and the risk of MAP introduction in the herd was found in 4 (9,29,33,35) of the 7 manuscripts. Daniels et al (29) found that farms with “more” wild rabbits had a higher mean probability (MP: 0.77, P = 0.018) of being infected than farms with “few” wild rabbits. The variable “access of wildlife to stored feed (farm mix)” (MP: 0.899, P = 0.028) was significantly associated with the probability of a farm reporting JD (29). Barrett et al (9) showed that co-grazing of heifers with sheep and/or goats was significantly associated with the detection of MAP in the univariate analysis (P < 0.05). Greig et al (35) showed that a past or current problem of paratuberculosis in cattle was significantly associated with paratuberculosis in the wild rabbit population on the farm. This study also showed the similarity (morphologically and by molecular genetic typing) between the isolates from both species. Çetinkaya et al (33) found that presence of farmed deer increased the risk of reporting disease with OR of 209 (95% CI: 13.47 to 3253) for 1 year of the 2 years studied and suggested that deer could be a reservoir of JD when they graze in the same pasture with cattle. However, these results must be considered with caution as only 0.4% of the farms in the study reported having deer.
Muskens (30) and Corn (34) did not find a significant relationship between the variables and the outcomes: “presence of sheep/goats grazing the same pasture versus the serological status of cattle,” nor between “wildlife on the farms versus the test-prevalence on livestock,” respectively.
Raizman et al (36) found a significant association between the use of pasture by cattle and the risk of contact between cattle manure and deer (OR: 5.4, 95% CI: 2.1 to 14.2) and also between cattle manure and rabbit (OR: 3.6, 95% CI: 1.3 to 9.5). However, they did not find a significant association between MAP in wildlife and infected herds.
Most of the references identified for the second question (14–17,19,37–45) showed a potential risk for cattle if wildlife or another domestic species were in close contact. However, these studies were not kept for the analysis because there was no measure of the risk and the emphasis was to report possible transmission.
Discussion
Introduction of cattle into a herd is a common, well-supported, and confirmed risk factor for herds to become infected with MAP (11). The principal reason to introduce/purchase animals is to expand herd size and many herds cannot do this by producing their own heifers. Another reason for introducing cattle is to acquire replacement stock, to restock a farm that suffers depopulation caused by another disease or to acquire an animal of high genetic value. The animal that is added could be infected and shedding MAP while showing no clinical signs. Due to the low number of certified MAP-free herds (12), the introduction/purchase of cattle from a safe source is a challenge and the primary preventive method should be to maintain a completely closed herd. Testing purchased cows before introduction is unrewarding because of the low sensitivity of tests on clinically healthy individual animals (12,46). Depending on the specific metric used in each study, the strength of association between purchase of replacement animals and risk of MAP infection, measured by ORs, was between 1.3 and 19.2. The wide range in ORs could be explained partly by differences in study design, in the tests used, and in the case definitions. Most studies had low to moderate internal validity, probably due to the same factors that affected the variability of association discussed. It is important to consider the validity in order to identify the impact of the findings of each study.
Daniels et al (29) did not find any association between the presence of the disease and number of cattle brought in. A possible explanation is that 15% of the farms took precaution by buying clean replacements; however, they did not define how a replacement was considered “clean.”
Two studies (28,30) had results that were contradictory to common knowledge. In Çetinkaya’s study (33), purchasing replacements privately was associated with an increased risk of reporting Johne’s disease compared with purchases from markets or from dealers. Purchasing replacements from known sources is considered a method to reduce the risk of introduction of disease. There is no clear explanation for this finding, other than the lack of consideration of the temporal sequence of events. If purchasing replacements from a known source is a management strategy to reduce disease burden, the consequence would be that already being affected by MAP caused the management to start purchasing low-risk animals through private agreements. In that case, the effect of buying from a known source could be wrongly associated with increased MAP prevalence. Tiwari et al (31) found that purchasing bulls was associated with a lower count of seropositive cows (protective factor). They hypothesized that farmers keeping bulls were applying other measures to reduce MAP seroprevalence. In other words, the variable could be a surrogate measure of some unmeasured preventive factor.
We found only 7 manuscripts that measured the association between presence of wildlife or other domestic species and presence of disease in cattle. However, many references (14–17,19,37–45) supported the widespread prevalence of MAP in wild species (rabbits, birds, feral cats, rats, raccoons, deer, opossums, mice, sheep, and goats). In several references, MAP was cultured from some species but no evidence of true transmission was measured. It is still unknown if wildlife are passive shedders of MAP or if they truly become infected. That wildlife are vectors and reservoirs of MAP seems to be undisputable (14) but the amplitude of the risk is not well-documented.
Raizman et al (36) estimated a significant association between the use of pasture by cattle and the risk of contact between cattle manure and both deer and rabbit. However, due to the low number of positive samples in wildlife in this study, an association between MAP in wildlife and infected herds was not significant.
In Stevenson’s study (19), paratuberculosis was diagnosed and the strain of MAP was isolated from different species on the same property. To assess the relative risk of transmission they suggest that a distinction between passive shedding and active transmission should be made. They also studied the possible contact between species. Unfortunately in this study, as in others, the authors did not measure the relationship between the presence of paratuberculosis in cattle and in wildlife. There is still much to investigate concerning the role and the degree of risk that wildlife pose to cattle.
We designed and applied a structured and repeatable search strategy, based on the most important concepts for each question and the keywords that best described them. The search strategy was initially wide and flexible to obtain all possible studies related to the subject and then was more restrictive in order to select the studies regarding the specific risk factors and criteria we were looking for. There were 3 main concepts for both questions for this review: “Paratuberculosis” “contamination” and “cattle,” and for each question the appropriate concepts to make the search more specific to the question were added.
The 2 questions formulated were related to the introduction of paratuberculosis into a herd. Even though the search for each question could be laborious, we considered that both were relevant for the objective of finding scientific evidence for the risk factors associated with the introduction of MAP into a herd.
We summarized the information qualitatively, due to the variability in the description of the risk factors, and to the high variability in expression of results. The appraisal remains subjective and potential bias may exist. The study appraisal classification (low — moderate — high) used by the authors did not take into account whether the results were statistically significant or not. The most common study design was the cross-sectional study. It provides low confidence in causal association and a moderate relevance to real situations (47). It remains a challenge to prove a true association between the outcomes (prevalence) and, for example, a management factor, which is a time-variant exposure. It is important to state that in this review we did not find studies that we could classify as having high internal and external validity. Although likely impractical, longitudinal observational studies of non-infected herds would be the most appropriate type of study design to identify true risk factors for introduction into a herd.
Another measure that should be included in a paratuberculosis program is the control of wildlife or other domestic animals on the farm and their interaction with cattle, although the importance of this as a risk factor remains to be determined.
Despite increased awareness of the disease and the fact that several countries are implementing control programs, there is still incomplete understanding of the epidemiology of the disease. Control programs must consider the inter-herd transmission factors in order to prevent the introduction of the disease.
In summary, this systematic review confirmed that the introduction or purchase of cattle is a risk factor for the introduction of paratuberculosis into a herd. This review did not find enough evidence to assess the importance of contact between wildlife/domestic animals and cattle as a risk factor for introduction of paratuberculosis into a herd. CVJ
Footnotes
The study was done in Saint-Hyacinthe, Québec, at the Faculté de Médecine Vétérinaire, Université de Montréal.
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
This study was supported by the Programme de Soutien à l’Innovation en Agroalimentaire du Ministère de l’Agriculture, des Pêcheries et de l’Alimentation du Québec.
References
- 1.Collins MT. Johne’s Disease (Paratuberculosis) In: David EA, Rings DM, editors. Food Animal Practice. 5th ed. St. Louis, Missouri: WB Saunders; 2009. pp. 65–69. [Google Scholar]
- 2.McKenna SLB, Keefe GP, Tiwari A, VanLeeuwen J, Barkema HW. Johne’s disease in Canada Part II: Disease impacts, risk factors, and control programs for dairy producers. Can Vet J. 2006;47:1089–1099. [PMC free article] [PubMed] [Google Scholar]
- 3.Waddell LA, Rajić A, Sargeant J, et al. The zoonotic potential of Mycobacterium avium spp. paratuberculosis: A systematic review. C J Public Health. 2008;99:145–155. doi: 10.1007/BF03405464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Behr MA, Collins DM. Paratuberculosis: Organism, Disease, Control. Wallingford, UK: CABI; 2010. p. 375. [Google Scholar]
- 5.Doré E, Paré J, Côté G, et al. Risk factors associated with transmission of Mycobacterium avium subsp. paratuberculosis to calves within dairy herd: A systematic review. J Vet Intern Med. 2012;26:32–45. doi: 10.1111/j.1939-1676.2011.00854.x. [DOI] [PubMed] [Google Scholar]
- 6.Whittington RJ, Windsor PA. In utero infection of cattle with Mycobacterium avium subsp. paratuberculosis: A critical review and meta-analysis. Vet J. 2009;179:60–69. doi: 10.1016/j.tvjl.2007.08.023. [DOI] [PubMed] [Google Scholar]
- 7.Nielsen SS, Toft N. Effect of management practices on paratuberculosis prevalence in Danish dairy herds. J Dairy Sci. 2011;94:1849–1857. doi: 10.3168/jds.2010-3817. [DOI] [PubMed] [Google Scholar]
- 8.Pithua P, Godden SM, Wells SJ, Oakes MJ. Efficacy of feeding plasma-derived commercial colostrum replacer for the prevention of transmission of Mycobacterium avium subsp paratuberculosis in Holstein calves. J Am Vet Med Assoc. 2009;234:1167–1176. doi: 10.2460/javma.234.9.1167. [DOI] [PubMed] [Google Scholar]
- 9.Barrett DJ, Mee JF, Mullowney P, et al. Risk factors associated with Johne’s disease test status in dairy herds in Ireland. Vet Rec. 2011;168:410. doi: 10.1136/vr.c6866. [DOI] [PubMed] [Google Scholar]
- 10.Orpin P, Duthie S, Grove-White DH. The use of targeted sampling and risk factor analysis to investigate the presence of Johnes disease in dairy herds. Cattle Pract. 2005;13:219–225. [Google Scholar]
- 11.Sibley RJ, Orpin PG. The risks of Johne’s disease entering and spreading within UK dairy herds. Cattle Pract. 2012;20:84–87. [Google Scholar]
- 12.Wells SJ, Wagner BA. Herd-level risk factors for infection with Mycobacterium paratuberculosis in US dairies and association between familiarity of the herd manager with the disease or prior diagnosis of the disease in that herd and use of preventive measures. J Am Vet Med Assoc. 2000;216:1450–1457. doi: 10.2460/javma.2000.216.1450. [DOI] [PubMed] [Google Scholar]
- 13.Greig A, Stevenson K, Perez V, Pirie AA, Grant JM, Sharp JM. Paratuberculosis in wild rabbits (Oryctolagus cuniculus) Vet Rec. 1997;140:141–143. doi: 10.1136/vr.140.6.141. [DOI] [PubMed] [Google Scholar]
- 14.Nugent G, Whitford EJ, Hunnam JC, Wilson PR, Cross ML, de Lisle GW. Mycobacterium avium subsp paratuberculosis infection in wildlife on three deer farms with a history of Johne’s disease. N Z Vet J. 2011;59:293–298. doi: 10.1080/00480169.2011.605747. [DOI] [PubMed] [Google Scholar]
- 15.Palmer MV, Stoffregen WC, Carpenter JG, Stabel JR. Isolation of Mycobacterium avium subsp paratuberculosis (MAP) from feral cats on a dairy farm with Map-infected cattle. J Wildl Dis. 2005;41:629–635. doi: 10.7589/0090-3558-41.3.629. [DOI] [PubMed] [Google Scholar]
- 16.Beard PM, Stevenson K, Pirie A, et al. Experimental paratuberculosis in calves following inoculation with a rabbit isolate of Mycobacterium avium subsp. paratuberculosis. J Clin Microbiol. 2001;39:3080–3084. doi: 10.1128/JCM.39.9.3080-3084.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kopecna M, Trcka I, Lamka J, et al. The wildlife hosts of Mycobacterium avium subsp paratuberculosis in the Czech Republic during the years 2002–2007. Vet Med-Czech. 2008;53:420–426. [Google Scholar]
- 18.Godfroid J, Boelaert F, Heier A, et al. First evidence of Johne’s disease in farmed red deer (Cervus elaphus) in Belgium. Vet Microbiol. 2000;77:283–290. doi: 10.1016/s0378-1135(00)00313-8. [DOI] [PubMed] [Google Scholar]
- 19.Stevenson K, Alvarez J, Bakker D, et al. Occurrence of Mycobacterium avium subspecies paratuberculosis across host species and European countries with evidence for transmission between wildlife and domestic ruminants. BMC Microbiol. 2009;9:212. doi: 10.1186/1471-2180-9-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Torgerson C. Systematic Reviews. London: Continuum International Publishing group; 2003. p. 102. [Google Scholar]
- 21.Sargeant JM, Rajić A, Read S, Ohlsson A. The process of systematic review and its application in agri-food public-health. Prev Vet Med. 2006;75:141–151. doi: 10.1016/j.prevetmed.2006.03.002. [DOI] [PubMed] [Google Scholar]
- 22.Khan KS, Kunz R, Kleijnen J, Antes G. Systematic reviews to support evidence-based medicine: How to review and apply findings of health-care research. London, UK: Royal Society of Medicine Press; 2003. [Google Scholar]
- 23.Sanderson RO, Beata C, Flipo RM, et al. Systematic review of the management of canine osteoarthritis. Vet Rec. 2009;164:418–424. doi: 10.1136/vr.164.14.418. [DOI] [PubMed] [Google Scholar]
- 24.Goodger WJ, Collins MT, Nordlund KV, et al. Epidemiologic study of on-farm management practices associated with prevalence of Mycobacterium paratuberculosis infections in dairy cattle. J Am Vet Med Assoc. 1996;208:1877–1881. [PubMed] [Google Scholar]
- 25.Obasanjo IO, Grohn YT, Mohammed HO. Farm factors associated with the presence of Mycobacterium paratuberculosis infection in dairy herds on the New York State Paratuberculosis Control Program. Prev Vet Med. 1997;32:243–251. doi: 10.1016/s0167-5877(97)00027-5. [DOI] [PubMed] [Google Scholar]
- 26.Ansari-Lari M, Haghkhah M, Bahramy A, Baheran AMN. Risk factors for Mycobacterium avium subspecies paratuberculosis in Fars province (Southern Iran) dairy herds. Trop Anim Health Prod. 2009;41:553–557. doi: 10.1007/s11250-008-9221-7. [DOI] [PubMed] [Google Scholar]
- 27.Cashman W, Buckley J, Quigley T, et al. Risk factors for the introduction and within-herd transmission of Mycobacterium avium subspecies paratuberculosis (MAP) infection off 59 Irish dairy herds. Ir Vet J. 2008;61:464–467. doi: 10.1186/2046-0481-61-7-464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hirst HL, Garry FB, Morley PS, et al. Seroprevalence of Mycobacterium avium subsp. paratuberculosis infection among dairy cows in Colorado and herd-level risk factors for seropositivity. J Am Vet Med Assoc. 2004;225:97–101. doi: 10.2460/javma.2004.225.97. [DOI] [PubMed] [Google Scholar]
- 29.Daniels MJ, Hutchings MR, Allcroft DJ, McKendrick IJ, Greig A. Risk factors for Johne’s disease in Scotland — The results of a survey of farmers. Vet Rec. 2002;150:135–139. doi: 10.1136/vr.150.5.135. [DOI] [PubMed] [Google Scholar]
- 30.Muskens J, Elbers ARW, van Weering HJ, Noordhuizen JPTM. Herd management practices associated with paratuberculosis seroprevalence in Dutch dairy herds. J Vet Med B. 2003;50:372–377. doi: 10.1046/j.1439-0450.2003.00697.x. [DOI] [PubMed] [Google Scholar]
- 31.Tiwari A, Vanleeuwen JA, Dohoo IR, et al. Risk factors associated with Mycobacterium avium subspecies paratuberculosis seropositivity in Canadian dairy cows and herds. Prev Vet Med. 2009;88:32–41. doi: 10.1016/j.prevetmed.2008.06.019. [DOI] [PubMed] [Google Scholar]
- 32.Nielsen SS, Toft N. Assessment of management-related risk factors for paratuberculosis in Danish dairy herds using Bayesian mixture models. Prev Vet Med. 2007;81:306–317. doi: 10.1016/j.prevetmed.2007.05.001. [DOI] [PubMed] [Google Scholar]
- 33.Cetinkaya B, Erdogan HM, Morgan KL. Relationships between the presence of Johne’s disease and farm and management factors in dairy cattle in England. Prev Vet Med. 1997;32:253–266. doi: 10.1016/s0167-5877(97)00028-7. [DOI] [PubMed] [Google Scholar]
- 34.Corn JL, Manning EJB, Sreevatsan S, Fischer JR. Isolation of Mycobacterium avium subsp. paratuberculosis from free-ranging birds and mammals on livestock premises. Appl Environ Microbiol. 2005;71:6963–6967. doi: 10.1128/AEM.71.11.6963-6967.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Greig A, Stevenson K, Henderson D, et al. Epidemiological study of paratuberculosis in wild rabbits in Scotland. J Clin Microbiol. 1999;37:1746–1751. doi: 10.1128/jcm.37.6.1746-1751.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Raizman EA, Wells SJ, Jordan PA, DelGiudice GD, Bey RR. Mycobacterium avium subsp paratuberculosis from free-ranging deer and rabbits surrounding Minnesota dairy herds. Can J Vet Res. 2005;69:32–38. [PMC free article] [PubMed] [Google Scholar]
- 37.Hutchings MR, Daniels MJ, Henderson D, Greig A. Potential wildlife to ruminant transmission routes for Mycobacterium avium subsp paratuberculosis. Proceedings of the Seventh International Colloquium on Paratuberculosis; June 11–14, 2002; Bilbao, Spain. 2003. pp. 363–367. [Google Scholar]
- 38.Daniels MJ, Hutchings MR, Greig A. The risk of disease transmission to livestock posed by contamination of farm stored feed by wildlife excreta. Epidemiol Infect. 2003;130:561–568. [PMC free article] [PubMed] [Google Scholar]
- 39.Daniels MJ, Hutchings MR, Beard PM, et al. Do non-ruminant wildlife pose a risk of paratuberculosis to domestic livestock and vice versa in Scotland? J Wildl Dis. 2003;39:10–15. doi: 10.7589/0090-3558-39.1.10. [DOI] [PubMed] [Google Scholar]
- 40.Daniels MJ, Henderson D, Greig A, Stevenson K, Sharp JM, Hutchings MR. The potential role of wild rabbits Oryctolagus cuniculus in the epidemiology of paratuberculosis in domestic ruminants. Epidemiol Infect. 2003;130:553–559. [PMC free article] [PubMed] [Google Scholar]
- 41.Judge J, Kyriazakis I, Greig A, Davidson RS, Hutchings MR. Routes of intraspecies transmission of Mycobacterium avium subsp. paratuberculosis in rabbits (Oryctolagus cuniculus): A field study. Appl Environ Microbiol. 2006;72:398–403. doi: 10.1128/AEM.72.1.398-403.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Judge J, Kyriazakis I, Greig A, Allcroft DJ, Hutchings MR. Clustering of Mycobacterium avium subsp. paratuberculosis in rabbits and the environment: How hot is a hot spot? Appl Environ Microbiol. 2005;71:6033–6038. doi: 10.1128/AEM.71.10.6033-6038.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Pieterse MC, Eisenberg SWF, Folmer GE, et al. Evidence of Mycobacterium avium subsp. paratuberculosis infection in Dutch farmed red deer. Tijdschr Diergeneeskd. 2010;135:886–890. [PubMed] [Google Scholar]
- 44.Machackova M, Svastova P, Lamka J, et al. Paratuberculosis in farmed and free-living wild ruminants in the Czech Republic (1999–2001) Vet Microbiol. 2004;102:225–234. doi: 10.1016/j.vetmic.2004.04.001. [DOI] [PubMed] [Google Scholar]
- 45.Pavlik I, Bartl J, Dvorska L, et al. Epidemiology of paratuberculosis in wild ruminants studied by restriction fragment length polymorphism in the Czech Republic during the period 1995–1998. Vet Microbiol. 2000;77:231–251. doi: 10.1016/s0378-1135(00)00309-6. [DOI] [PubMed] [Google Scholar]
- 46.More SJ, Sergeant ES, Strain S, Cashman W, Kenny K, Graham D. The effect of alternative testing strategies and bio-exclusion practices on Johne’s disease risk in test-negative herds. J Dairy Sci. 2013;96:1581–1590. doi: 10.3168/jds.2012-5918. [DOI] [PubMed] [Google Scholar]
- 47.Dohoo I, Martin W, Stryhn H. Veterinary Epidemiologic Research. 2nd ed. Charlottetown, Prince Edward Island: VER Inc; 2010. [Google Scholar]
- 48.Wells SJ, Whitlock RH, Wagner BA. Sensitivity of test strategies used in the Voluntary Johne’s Disease Herd Status Program for detection of Mycobacterium paratuberculosis infection in dairy cattle herds. J Am Vet Med Assoc. 2002;220:1053–1057. doi: 10.2460/javma.2002.220.1053. [DOI] [PubMed] [Google Scholar]
- 49.Nielsen SS. Variance component of an enzyme-linked immunosorbent assay for detection of IgG antibodies in milk samples to Mycobacterium avium subspecies paratuberculosis in dairy cattle. J Vet Med B Infect Dis Vet Public Health. 2002 Oct;49:384–387. doi: 10.1046/j.1439-0450.2002.00592.x. [DOI] [PubMed] [Google Scholar]