Tropical Animal Health and Production

, Volume 41, Issue 4, pp 553–557

Risk factors for Mycobacterium avium subspecies paratuberculosis in Fars province (Southern Iran) dairy herds

Authors

    • Department of Food Hygiene and Public Health, School of Veterinary MedicineShiraz University
  • Masoud Haghkhah
    • Department of Pathobiology, School of Veterinary MedicineShiraz University
  • Ayatollah Bahramy
    • Department of Pathobiology, School of Veterinary MedicineShiraz University
  • Amir Mansour Novin Baheran
    • Department of Pathobiology, School of Veterinary MedicineShiraz University
Original Paper

DOI: 10.1007/s11250-008-9221-7

Cite this article as:
Ansari-Lari, M., Haghkhah, M., Bahramy, A. et al. Trop Anim Health Prod (2009) 41: 553. doi:10.1007/s11250-008-9221-7

Abstract

A cross-sectional study was conducted from March to August 2006 in dairy herds in Fars province, southern Iran to determine the herd-level risk factors for infection with Mycobacterium avium subspecies paratuberculosis (MAP). Statistical analysis using multivariable logistic regression showed that contamination of udders of periparturient cows with manure (OR = 6.4, P = 0.02) and history of having suspected cases of Johne's disease in the herd (OR = 6.7, P = 0.04) were significantly associated with the herd infection status. No relationship between breed, herd size and other management practices with the infection status of the herd were found in this study. Implementing high sanitary measures in the farm, particularly with respect to manure handling and cleaning could be considered as one of the important aspects in controlling disease in the region as well as in the future educational effort.

Keywords

IranJohne’s diseaseMycobacterium avium subspecies paratuberculosisRisk factor

Introduction

Johne’s disease is a chronic disease of ruminants worldwide caused by Mycobacterium avium subsp. paratuberculosis (MAP). The Johne’s disease in the dairy community has long been considered as an economically-important disease because of its negative impact on milk production and the overall economic health of dairies (Benedictus et al. 1987; Beaudeau et al. 2007; Gonda et al. 2007). This disease has gained notoriety in recent years because of concern about a possible relationship to Crohn’s disease, an intestinal disorder in humans (Greenstein and Collins 2004; Gaya et al. 2004; Romero et al. 2005).

A primary method for transmission of MAP is through fecal contamination of calf environment including contamination of milk and feed (Sweeney 1996). In addition pooled milk or colostrums can be important sources of infectious material for newborn calves. Therefore, management of calving area and parturient cows and care of growing newborn calves are among general practices which are recommended for the prevention of Johne’s disease (Russiter and Burhans 1996).

From the epidemiological point of view and for implementation of good control program, it is recommended to find associated risk factors in each area. This is due to the complexity of the causal web for transmission of infection; a factor which is not so important in one area may be an important causal complement in another one and vise versa. The aim of the present study, therefore, was to identify possible association between herd management practices and MAP infection status of the herds in Fars province, southern Iran. Herds’ infection status was determined based on IS900-PCR on bulk- tank milk samples in a concurrent study.

Materials and methods

Study population

This is a cross-sectional study which was conducted in Fars province, southern Iran. Fars province plays an important role in dairy industry in Iran. Industrial dairy herds does not distributed evenly in the province. There are three main districts (Shiraz, Marvdasht, and Sepidan) in the province which comprised more than 80% of all dairies in the region. Therefore these districts were considered to be nearly representative of the province and the study population was selected from there. There are about 120 industrial dairy herds in this region. Selection of herds was accomplished by Jihad Agriculture Organization of Fars province which has the name of owners and location of each herd in the province. Overall, 110 herds were included in the study, because a few herds were not interested in participating in the study. The herd-level prevalence of M. paratuberculosis infection in these herds was 11% (95% CI: 5–17%) which was determined by IS900-PCR assay on bulk-tank milk samples (Haghkhah et al. 2008).

A structured questionnaire was used to collect general information about the herd such as breed and herd size as well as herd management practices. Most questions were directed to evaluate the management practices that thought to be associated with increased prevalence and are important in the transmission of MAP in specific production areas: calving, pre-weaned calves, post-weaned calves, heifers and adult cows. Also questions about history of suspected cases of Johne’s disease (chronic diarrhea and weight loss despite normal appetite that does not respond to treatment), and source of replacement cows were included. Most factors and questions were compiled from the related literature; also all were designed as closed questions. A brief description of questions is provided in Table 1.
Table 1

Distribution of risk factors for Mycobacterium paratuberculosis infection in 110 dairy herds in Fars province, southern Iran based on IS900-PCR in balk-tank milk samples (March to August 2006)

Risk factors

PCR positive

PCR negative

N

%

N

%

Area*

Shiraz

5

8.6

53

91.4

Marvdasht

3

8.6

32

91.4

Sepidan

4

23.5

13

76.5

Breed*

Holstein

8

8.8

83

91.2

Cross-bred (Holstein-Indigenous)

4

21.1

15

78.9

Calves stay with their dams more than 3 hrs

Yes

1

10

9

90

No

11

11

89

89

Contamination of adult cow’s feed or water by manure

yes

9

10.2

79

89.8

No

3

14.3

18

85.7

Contamination of heifer’s feed or water by manure

yes

10

12.2

72

87.8

No

1

3.7

26

96.3

Contamination of post-weaned calf’s feed or water by manure*

Yes

7

20.6

27

79.4

No

5

6.6

71

93.4

Contamination of udders of perparturient cows with manure*

yes

4

44.4

5

55.6

No

8

7.9

93

92.1

Herd size

<50

8

13.3

52

86.7

50–99

3

7.7

36

92.3

>100

1

10.0

9

90.0

History of Johne’s disease in the herd*

Yes

3

37.5

5

62.5

No

9

8.8

93

91.2

Keep post-weaned calves near adult cows*

Yes

9

14.5

53

85.5

No

3

6.3

45

93.8

Pooled colostrums fed to calves

Yes

3

11.5

23

88.5

No

9

10.7

75

89.3

Purchase replacement animals

Yes

4

11.4

31

88.6

No

8

10.7

67

89.3

Separate area for calving

yes

7

11.9

52

88.1

No

5

9.8

46

90.2

Unpasteurized milk fed to calves

Yes

12

10.9

98

89.1

No

0.0

0.0

0.0

0.0

*Significant factors in the univariable analysis (P < 0.2)

Completion of the questionnaires was performed by two authors of the study during farm visits. Questions about general herd information were answered directly by herd owners or another person of farm staff. Because most owners have difficulties in being objective when they have to judge and answer the questions, most of the other questions were answered based on observation at the time of interview.

Statistical analysis

The unit of statistical analysis was herd. To evaluate associations between suggested risk factors and herd level status for M. paratuberculosis infection (outcome variable), univariable and multivariable logistic regression analysis were used. Factors with P-value <0.2 in the univariable analysis were included in the multivariable logistic regression analysis. Multicollinearity amongst putative risk factor variables was assessed using Spearman’s Rank correlation coefficients. Also correlation between each independent variable and linear combination of other variables was investigated using variance inflation factor (VIF). When computed correlation coefficients were higher than 0.35, one of those correlated variables with less P-value in the univariable analysis was included in the multivariable model. Also, variables with VIF higher than 1.3 were excluded from the multivariable model. The multivariable logistic regression model was constructed using a step-wise backward elimination approach. Those variables with Wald statistic values with P < 0.05 in the final model were considered as statistically significant. All statistical analysis was conducted using SPSS software (Version 11.5).

Results

With univariable analysis and after removing the correlated variables, six factors were selected based on P value <0.2 (Table 1) and were included in the multivariable step-wise logistic regression analysis. In the final logistic model, two out of six factors were significant at the 0.05 level. No significant interaction was detected in the regression model. The Wald statistic, standard error, odds ratios and the 95% confidence level for the factors are provided in Table 2. Herds with low sanitation of periparturient cows as measured by high contamination of their udders with manure had 6.4 times the odds of being positive for MAP infection. Also herds with history of having suspected cases of Johne’s disease had 6.7 times more likely to be positive than herds without such history. Herd size, breed, having separate area for calving, feeding calves with pooled colostrums and keeping calves more than three hours with their dams were among factors which were not significant in the statistical analysis. Also, location of herds (geographical area) was not significant in the final model.
Table 2

Multivariable logistic regression analysis of risk factors for Mycobacterium paratuberculosis infection in 110 dairy herds in Fars province, southern Iran based on IS900-PCR in balk-tank milk samples

Risk factor

Wald statistic

SE

OR

95% CI

P

Contamination of udders of periparturient cows with manure

Yes

5.18

0.81

6.38

1.29–31.49

0.02

No

-

-

1.00

-

-

History of Johne’s disease in the herd

Yes

3.91

0.96

6.70

1.02–44.16

0.04

No

-

-

1.00

-

-

Discussion

In Iran, the first report of paratuberculosis comes back to about 40 years ago by Talachian (1965) which showed that the origin of infection was imported animals. There are several other reports from various parts of the country about the occurrence of disease and its undesirable consequences in cattle, sheep, and goats (Anzaby et al. 2006; Pourjafar and Badiei 2005; Khodakaram Tafti and Rashidi 2000). However, to the best of the authors’ knowledge, there is no study about risk factors of MAP infection in the country.

Several practices in different aspects of production areas that thought to be important in the prevalence and transmission of MAP were evaluated in this study. High probability of udders of periparturient cows being contaminated by manure was significantly associated with infection status of herd. This is consistent with reports from other studies which showed that increased probability of contact of calves to feces of adult cows (Obasanjo et al. 1997) or extent of manure build-up in the calving area (Berghaus et al. 2005) may be among important risk factors for infection with MAP. It could be plausible because calves are most susceptible group to MAP infection, and low sanitation of udders may potentially increase the possibility of manure ingestion by neonate calf during or immediately after birth.

History of having suspected cases of Johne’s disease in the herd was also associated with an increased risk of being positive for MAP infection. Similar results were reported in studies in Michigan and New York state dairy herds, USA (Johnson-Ifearulundu and Kaneene 1998; Obasanjo et al. 1997). Because Johne’s disease is generally a subclinical infection, plausibility of this relationship is relatively apparent.

There was no difference between breeds, including Holsteins and cross-bred regarding the infection with MAP. It is generally accepted that indigenous cows are relatively more resistant to endemic diseases. According to the present results, for Johne’s disease which is originally an exotic disease, no difference between Holsteins and cross-bred herds was observed. On the other hand, it is noteworthy that distinction between cross-bred and Holsteins may be problematic for the farmers. Indeed, sometimes dairy cattle in industrial farms are referred to as “Iranian Holsteins”, because they are considered Holsteins with varying degree of genetic purity due to cross-breeding with indigenous cows. Taken together, for more accurate comparison of breeds, infection status of individual cows, and considering their genetic purity, also the severity of clinical course of disease is needed.

Several other factors that would be expected to be potential risk factors for MAP infection were not significant in the present study. One reason may be the cross-sectional design of the study. Changes in the management practices that have been implemented in response to previous clinical cases in the farm could potentially blurred their actual association with MAP infection status. On the other hand, considering the various environmental, nutritional and management conditions over the world, there is possibility that some factors may work as important causal complement for MAP infection in one region and might not be so important in another region.

In conclusion, history of suspected cases of Johne’s disease and contamination of udders of periparturient cows with manure, both indicate a higher likelihood that a herd is infected with M. paratuberculosis. Implementing high sanitary measures in the farm, particularly with respect to manure handling and cleaning could be considered as one of the important aspects in controlling disease in the region as well as in the future educational efforts.

Acknowledgments

The authors would like to thank the herd owners for their help and cooperation in collecting management information. This study was supported by Grant No. 84-VE-1780_C308 Shiraz University, Fars, Iran.

Copyright information

© Springer Science+Business Media B.V. 2008