Materials and Methods

The present study was part of a larger study designed to compare failure of passive transfer rates, pre-weaning health, and risk of MAP infection in calves fed raw bovine colostrum versus a commercially available plasma-derived colostrum replacer^a at birth. Detailed colostrum feeding protocols and study design for the previous study have been described.^17,20

Herd selection and animals—Holstein calves born on 12 commercial dairy farms between July and October 2003 were used in the study. Herds were located in western Minnesota (n = 9 farms) and eastern Wisconsin (3 farms), and participation in this study was entirely voluntary. Criteria for herd inclusion were producer willingness to comply with study protocols, participation in the Minnesota or Wisconsin Dairy Herd Improvement Association milk production records system, and evidence that paratuberculosis was endemic in the herd. Endemic status was initially determined on the basis of a history of culling cattle because of paratuberculosis during the year preceding the start of the study (ie, 2002), although confirmatory testing to determine herd infection status with MAP was subsequently performed, as described.¹⁷ Before onset of this study, farm managers were asked to complete a questionnaire describing their routine colostrum manage ment practices and general herd information. Approval for the conduct of this study was granted by the Institutional Animal Care and Use Committee of the University of Minnesota.

Experimental design and calf feeding protocols—A prospective cohort design was used to estimate the risk of subclinical MAP infection in calves fed colostrum with MAP DNA (ie, exposed group), compared with the risk of infection in calves fed colostrum without MAP DNA (ie, unexposed group) at birth. Biological samples were obtained, and data collection was completed in December 2007.

Calves were separated from their dams within 30 to 60 minutes after birth and denied the opportunity to suckle their dams. Each calf was then fed colostrum originating from its dam. However, when a sufficient quantity of colostrum was not available from their dams, calves were fed refrigerated colostrum from other cows.

Feeding protocols for colostrum varied among study herds with respect to the quantity fed and frequency of feeding. For example, calves born in 11 of the 12 participating herds were fed 3.8 L of raw colostrum at the first feeding, whereas calves born in the remaining herd were fed 1.9 L of raw colostrum during the first feeding. Calves born in 5 herds were routinely provided a single meal of raw colostrum followed by a commercial milk replacer thereafter. In the other 7 herds, calves were routinely provided a second meal consisting of 1.9 L of raw colostrum collected from their dams or originating from the same source as the first colostral meal within 8 to 12 hours after the first colostral feeding. Preweaning diets (up to 56 days of age) for all calves consisted of commercial milk replacer and unrestricted amounts of water and calf starter pellets.

Preweaning sample and data collection—For each enrolled calf, calf managers recorded the identification of the dam, identification of the calf, birth date, number of calves born (singletons or twins), calving score, and interval from birth until feeding of colostrum.

Paired 20-mL samples of the colostrum fed to each calf at the first feeding were collected into sterile sample tubes that were subsequently labeled with the identity of the calf and farm prior to freezing at −20°C. The samples were stored for approximately 4 years at the College of Veterinary Medicine, University of Minnesota, prior to analysis to determine contamination with MAP. The intent was to establish whether calves were exposed to MAP through ingestion of raw colostrum. The colostrum samples were shipped to the USDA Agricultural Research Service National Animal Disease Center in Ames, Iowa, for further processing and analysis.

Postweaning samples and data collection—Following the feeding of colostrum, all calves were raised under the same management conditions (eg, nutrition, housing, and environment) within any given study farm or professional heifer grower's facility. All calves were weaned at approximately 56 days of age and monitored until adulthood. Data on calves were not collected by the investigators between weaning and the first testing event at 30 months of age. However, records of deaths, sale of animals, or culling events during this period were retrieved from commercially available dairy management software^b used in participating farms. Alternately, some of the death and culling events were accessed from records maintained manually by the producers. Despite this effort, some heifers were lost from the study during this period. These heifers were censored at the time of weaning. There were no losses to follow-up in the period between the first testing events (30 months of age) until conclusion of the study (54 months of age). Prior to the start of testing of the adult study cows for MAP infection, producers completed a questionnaire describing details of their routine paratuberculosis management programs.

After entering the lactating herd as adults, study animals were tested for MAP infection at approximately 30, 42, and 54 months of age. During each sampling event, 10 mL of blood was collected from the coccygeal vein by use of a 20-gauge, 1-inch needle into a 16 × 100-mm blood collection tube.^c Approximately 10 g of feces was collected manually from the rectum by use of a disposable plastic examination sleeve and stored in sterile plastic sample containers prior to shipment to the Minnesota Veterinary Diagnostic Laboratory located at the College of Veterinary Medicine, University of Minnesota, for processing and testing. Adult cow lactation records, including calving dates, milk production, and dates of death or culling events during this period, were retrieved from commercially available dairy management software^b used on participating farms.

Sample testing procedures—Colostrum samples were analyzed for MAP DNA at the USDA Agricultural Research Service National Animal Disease Center. Testing of serum samples and bacterial culture of feces were conducted at the Minnesota Veterinary Diagnostic Laboratory.

PCR analysis of colostrum—Mycobacterium avium subsp paratuberculosis DNA was extracted by adding an equal volume of phenol, chloroform, and isoamyl alcohol (25:24:1)^d to 1 mL of the colostrum samples. Details of this analysis have been reported elsewhere.²¹ The PCR protocol used 2 runs with a conventional thermal cycler^e followed by real-time PCR.^f

Initial sample runs were completed in triplicate in 96-well plates according to the following protocol: 1 cycle at 94°C for 5 minutes, followed by 20 cycles at 94°C for 45 seconds, 58°C for 1 minute, and 72°C for 2 minutes, and a final extension cycle at 72°C for 7 minutes. The primer sequences for the initial amplification were 5′-GCACGGTTTTTCGGATAACGAG-3′ (forward primer) and 5′-TCAACTGCGTCACGGTGTCCTG-3′ (reverse primer).

Amplification was performed by use of real-time PCR^f and included a fluorophore 6-carboxyfluorescein–labeled probe specific for the ISMAP02 target sequence for the semiquantitative evaluation of the test samples. Primers nested within the first set included 5′-GGATAACGAGACCGTGGATGC-3′ (forward primer) and 5′-AACCGACGCCGCCAATACG-3′ (reverse primer) for this second amplification, yielding a 117-bp product. One microliter of DNA from the first amplification was added to a reaction mixture consisting of PCR mastermix,^g ultrapure distilled water (DN-ase- and RNase-free), 0.05μM primers, and 0.05μM of the fluorophore 6-carboxyfluorescein–labeled probe (5′-/56-FAM/CAACCCGCACGCTG/3BHQ-1/-3′). A standard was constructed by amplifying the ISMAP02 target from MAP strain K-10 genomic DNA and cloning it into the topoisomerase I expression vector^h followed by transformation in Escherichia coli. The insert was analyzed for accuracy following plasmid digestion with EcoRI and verification of size on a 4% agarose gel. Further verification of the cloned insert was conducted by sequencing the product by use of a commercial DNA analyzerⁱ after labeling the product with a cycle-sequencing kit.^j Primers used for sequencing were M13F (5′-CGTTGTAAAACGACGGCCAGT-3′; forward) and M13R (5′-CAGGAAACAGCTATGAC-3′; reverse). Optimal concentrations of the plasmid containing the gene of interest used to construct the real-time PCR standard curve ranged from 1 to 100 ng. Real-time PCR conditions for the amplification of test samples and standards were 1 cycle at 50°C for 2 minutes, 1 cycle at 95°C for 10 minutes, 40 cycles at 95°C for 25 seconds, and 60°C for 1 minute. Positive samples were plotted in the form of sigmoid curves on 2-D grids, with the x-axis representing the PCR cycle number and the y-axis representing the relative fluorescence of the signal. Sample runs with mean threshold cycles in the range of 15 to 30 cycles were considered positive.

Serum ELISA for anti-MAP antibodies—Blood samples were tested for antibodies against MAP by use of a commercially available serologic kit.^k Results were indicated by optical density values. For each sample, values for anti-MAP antibodies were reported as a sample-to-positive ratio. Samples with a sample-to-positive ratio < 0.25 were classified as having negative results, whereas those with a sample-to-positive ratio ≥ 0.25 were classified as having positive results. In a validation study,²² sensitivity of this test ranged from 15.5% to 88.1%, depending on the clinical stage and amount of MAP fecal shedding of each infected cow.

Bacterial culture for MAP in fecal samples—The method for the cultivation of MAP in feces used in the present study has been described in detail.^23,24 Briefly, sediments derived from the fecal samples were plated onto Herrold egg yolk agar slants^l containing mycobactin J and an antimicrobial (0.1 mL of 50 mg amphotericin B) and incubated at 37°C for 3 weeks. At approximately 6 weeks after inoculation, the agar slants were examined weekly for growth of MAP. On the basis of total colony counts per gram of fecal sample tested, results were reported as negative (0 colonies/tube) or positive (1 to ≥ 100 colonies/tube).

Definition for subclinical MAP infection—Subclinical MAP infection of study animals was indicated on the basis of the following case definitions: a positive ELISA result for MAP only, a positive result of bacterial culture of feces for MAP only, or a positive ELISA result and bacterial culture of feces result for MAP based on a parallel diagnostic test interpretation criterion.²⁵ Although some study cows had clinical signs consistent with paratuberculosis over the study follow-up period, clinical signs were not considered as part of the MAP infection case definition. Moreover, study cows that eventually had clinical signs consistent with paratuberculosis already had positive results to either the ELISA or bacterial culture of feces diagnostic tests. Although some cows tested positive for MAP on more than 1 occasion over the study follow-up period at approximately 30, 42, and 54 months, only the first MAP-positive event on the serum ELISA and bacterial culture of feces tests was considered as the outcome for the purpose of data analysis.

Statistical analysis—Descriptive and multivariable analyses were performed by use of standard statistical software,^m and values of P ≤ 0.05 were considered significant. The proportion of cows with positive test results for MAP at each sampling event (approx 30, 42, and 54 months of age) and the overall proportion positive for MAP (all months combined) were summarized. Unconditional associations between feeding MAP DNA-positive versus -negative colostrum and a subclinical MAP-positive test outcome among study cohorts were evaluated by use of the χ² statistic at each sampling event (approx 30, 42, and 54 months of age).

For each cow in the study, months of observation was calculated from the time of enrollment (calf birth date) to the sampling date when the first positive test result for MAP was established or when the study concluded at approximately 54 months of follow-up monitoring, at which time all surviving cows were censored. Cows that died or were culled for any reason during the 54-month follow-up period were censored at the date of their death or culling. Cows lost during the follow-up monitoring period between weaning and the first testing event (approx 30 months of age) were assumed to have exited the study at weaning and were therefore censored from the analysis at date of weaning (approx 56 days after birth).

Weibull hazard regression models (3 models) were fit to the data to evaluate the association between feeding MAP DNA-positive versus -negative colostrum and the hazard of a positive MAP diagnosis (as determined on the basis of the serologic ELISA, bacterial culture of feces, or both with parallel test interpretation), and herd effect was adjusted for by inclusion of an inverse-Gaussian shared-herd frailty term.^26,27 Weibull distributions provide monotonically increasing or decreasing shapes of the hazard function determined by an estimated parameter, ρ, with a value of ρ = 1 implying a constant hazard function. Values of ρ < 1 are suggestive of monotonically decreasing hazard function, whereas values of ρ > 1 are indicative of a monotonically increasing hazard function.²⁶

The Weibull distribution was chosen to model the hazard of MAP infection in exposed versus unexposed groups for 2 reasons. First, it provided a good fit to the data in comparison with other evaluated hazard distributions (eg, exponential, Gompertze, log normal, and log logistic) on the basis of Akaike information criteria values (ie, the best-fitting model is the one that yields the smallest Akaike information criteria value) computed for each model with the exposure variable as the only predictor. Second, the long incubation period for MAP infections in cattle (assumed to be ≥ 2 years) meant that with the passage of time, the hazard rate of detecting infection increases. Because the diagnostic tests used in the present study have sensitivity that increases with progression of MAP infection from a subclinical to clinical state,^22,28 in the authors' opinion, the Weibull hazard distribution provided the most suitable description for the natural course of MAP infection.

In addition, the choice between γ and inverse-Gaussian distributions to model the shared-herd frailty effect was made on the basis of their respective fit to the data, with a better fit indicated by a higher maximum log likelihood of each full model (ie, models with the herd frailty terms included) and how well the shape of the frailty distributions fit the theoretical description of the herd effect. The γ frailty effect assumes that with passage of time, the effect of model variables on the population hazard decreases in favor of the frailty effect (ie, the herd effects would completely dominate in the long term). Conversely, an inverse-Gaussian frailty effect assumes that the effects of model variables decrease with time, although they never completely disappear.²⁶ In this analysis, the inverse-Gaussian shared-herd frailty distribution resulted in models with a better fit than that for the γ shared-herd frailty distribution (log likelihood for the serologic ELISA, −34.7 vs −33.3 for inverse-Gaussian vs γ shared-herd frailty, respectively; bacterial culture, −49.2 vs −48.8 for inverse-Gaussian vs γ shared-herd frailty, respectively; and for 1 or both diagnostic tests [parallel interpretation for ELISA and bacterial culture], −51 vs −50.5 for inverse-Gaussian vs γ shared-herd frailty, respectively). Thus, the inverse-Gaussian shared-herd frailty models were preferred.

Finally, the goodness of fit of the final models was evaluated by generating plots of cumulative hazard function against Cox-Snell residuals with a cumulative hazard approximating a 45° line with a zero intercept considered as suggestive of a good fit of the model to the data.²⁷ A visual assessment of the shapes of plots of cumulative hazard function against Cox-Snell residuals suggested that the final models provided good fits for the data.

Results

Herd description, paratuberculosis management, and herd MAP prevalence—Characteristics of participating herds with respect to size, annual milk production, number of calves enrolled per herd, time to first feeding of colostrum, method of colostrum feeding, number of colostrum feedings, site at which the calves were raised, and type of calf housing during the pre- weaning period have been described.²⁰

Briefly, mean number of cows in participating herds was 650 (range, 190 to 1,550). Fifty-two percent of dairy herds in Minnesota and 54% of dairy herds in Wisconsin are within this range for number of cows in herd.ⁿ Mean milk production (rolling herd mean) for each farm was 11,538 kg/y (range, 9,761 and 13,620 kg/y). Bulk-tank somatic cell count ranged from 180,000 to 280,000 cells/mL (median, 240,000 cells/mL).

Median interval from birth to feeding of the first colostral meal was 1 hour (range, 0.17 to 9 hours). Colostrum was bottle-fed to calves in 6 herds, whereas in the remaining 6 herds, colostrum was fed to calves via an esophageal tube feeder.

Calves were raised on-site at 6 farms, whereas the other 6 farms transported young (1 to 3 days old) calves to be raised off-site at the facilities of 1 of 2 professional heifer growers. Lactating cows were housed in free stalls at all participating dairies. Housing for calves raised onsite comprised individual calf hutches, whereas calves raised offsite by the professional heifer growers were commingled with calves from other nonparticipating herds and housed in separate pens inside barns. Calves raised by the professional heifer growers were transported back to their original farms as pregnant heifers at approximately 22 to 24 months of age.

All participating herds were infected with MAP, with the prevalence of infection differing among herds. Estimated true MAP prevalence for each herd ranged from < 1% to 64% with a median of 8%. However, MAP was not present in the environmental fecal samples collected from the professional heifer growers' facilities, suggesting that the facilities were MAP free and calves raised there were likely not exposed to MAP during that period.

Eight of the 9 farms in Minnesota were participants in the Minnesota Voluntary paratuberculosis Control Program. Only 1 of the 3 farms enrolled from Wisconsin reported participation in a similar voluntary paratuberculosis control program. Although MAP-testing strategies varied among study herds, they were not mutually exclusive because some of the herds reported use of > 1 MAP-testing strategy. In 5 herds, only cattle that had clinical signs of paratuberculosis were tested, whereas only cattle purchased from external sources were tested in 1 herd. In 4 herds, cattle were tested only at the end of lactation. Whole-herd testing for MAP was reported for 2 herds. In 5 herds, the decision to cull a cow suspected of being infected with MAP was made on the basis of a positive diagnostic test result and detection of clinical signs of paratuberculosis. In 6 herds, culling decisions were made on the basis of only clinical signs of MAP, whereas MAP-suspect cows were culled solely on the basis of positive diagnostic test results for MAP in only 1 herd.

Descriptive analysis—A flow chart representing number of calves available for testing at each time point was made (Figure 1). The sample of interest was the subgroup of heifer calves fed raw colostrum (n = 261) at birth, although only 205 calves were included in the present analysis; 56 calves were excluded because colostrum samples were not collected for analysis as intended. Of the 205 colostrum samples tested, 34% (69/205) were positive for MAP DNA versus 66% (136/205) that tested negative for MAP DNA. Therefore, 34% (69/205) of the calves that ingested colostrum that tested positive for MAP DNA were classified as exposed, and 66% (136/205) of the calves that ingested colostrum that tested negative for MAP DNA were categorized as unexposed.

View Full Size

Flow diagram illustrating the numbers of cattle included in analysis or censored at approximately 30, 42, and 54 months of a study performed to estimate the risk of subclinical MAP infection in 205 heifer calves that ingested MAP DNA–positive raw colostrum, compared with calves that ingested MAP DNA–negative raw colostrum, after birth. Note that the number tested at each time point is the difference between number of cows censored between each testing time point and the number tested during the previous testing cycle.

Citation: American Journal of Veterinary Research 72, 11; 10.2460/ajvr.72.11.1456

• Download Figure
• Download figure as PowerPoint slide

Overall, 14 of 205 (7%) cows had positive ELISA results, including 10 of 136 (7%) in the unexposed group and 4 of 69 (6%) in the exposed group. Similarly, 24 of 205 (12%) cows had positive results for MAP on the basis of results of bacterial culture of feces, including 16 of 136 (12%) in the unexposed group and 8 of 69 (12%) in the exposed group. When the positive outcomes for the ELISA and bacterial culture tests were combined via a parallel test interpretation criterion, the overall proportion of cows with positive test results for MAP was 13% (26/205), with 13% (18/136) in the unexposed group versus 12% (8/69) in the exposed group (Table 1).

Multivariable analysis—Predicted hazard functions for the final models comparing exposed versus unexposed groups were plotted (Figure 2). Hazards close to zero were observed up to approximately 30 months, reflecting the duration of the follow-up period when no MAP testing was performed. Hazard of MAP infection gradually increased in both groups up to approximately 54 months. However, although the increase was more rapid in the exposed group, suggesting that cows in the exposed group had an apparently greater risk of developing MAP infection with the passage of time (or alternatively that exposed cows tested positive for MAP much earlier), in comparison with the unexposed group, there were no significant differences between groups (exposed vs unexposed).

View Full Size

Weibull hazard functions for time to MAP-positive test outcome for MAP infection on the basis of serum ELISA (A), bacterial culture of feces (B), and 1 or both diagnostic tests (parallel interpretation; C) in exposed calves (solid lines) or unexposed calves (dotted lines).

Citation: American Journal of Veterinary Research 72, 11; 10.2460/ajvr.72.11.1456

• Download Figure
• Download figure as PowerPoint slide

Weibull regression models adjusted for the shared-herd frailty effect were determined on the basis of records for the 205 heifer calves (Table 2). Estimated shape parameters (ρ) for the models were as follows: ELISA, ρ = 3.88; bacterial culture, 3.92; and both diagnostic tests (parallel interpretation for ELISA and bacterial culture), 4.02. Shape parameter values suggested a monotone increasing hazard of MAP infection for the exposed and unexposed groups. Shared-frailty terms for the herd variable were significant in all 3 models, suggesting the existence of true herd cluster effect.

Findings from the final Weibull models indicated that feeding MAP DNA–positive versus MAP DNA–negative raw colostrum at birth did not significantly increase the hazard of MAP infection in cows as indicated by the ELISA, bacterial culture, and either 1 or both diagnostic test (parallel interpretation) outcome models (Table 2).

Discussion

To the authors' knowledge, this is the first study undertaken to determine the association between ingestion of MAP DNA–positive colostrum versus MAP DNA–negative colostrum and risk of transmission of MAP in calves at birth by testing the exposed calves for subclinical MAP infection over time.

Strengths of this study included its longitudinal cohort design and the large number of participating herds (n = 12), which ensured inclusion of diverse herd populations. Participating herds were confirmed MAP-infected herds that provided the ideal populations needed for addressing the present study objective.

Cows that ingested MAP DNA–positive colostrum (ie, exposed calves) as calves in the present study were expected to have a higher risk of subclinical MAP infection than the unexposed cows (ie, those that ingested MAP DNA–negative colostrum). This was premised on the understanding that MAP DNA was a surrogate marker for the presence of viable MAP present in the samples of colostrum fed, potentially resulting in MAP infection upon ingestion by calves after birth. There was, however, no significant difference in the risk of subclinical MAP infection between the exposed group and the unexposed group, suggesting that there is little to no added risk of subclinical MAP infection associated with ingesting MAP DNA–positive colostrum by calves.

Despite the present study's strengths, there were a number of limitations that need to be considered. First, in 7 of the 12 herds, a second meal of colostrum collected from their respective dams or originating from the same source as the first colostral meal was routinely provided within 8 to 12 hours after the first meal was fed. However, samples of colostrum fed to the calves at the second feeding were not tested for MAP DNA, so this second feeding was potentially an additional source of exposure to MAP.

Second, the investigators were unable to confirm dam status for MAP infection (specifically MAP fecal shedding status) prior to or even after the birth of each enrolled calf. Evidence for possible in utero MAP transmission has been reported,^29–31 with an estimated 9% of fetuses carried by subclinically infected dams being infected by MAP at any single time, through in utero exposure.³² In addition, MAP-infected dams are more likely to excrete MAP in colostrum than their MAP-negative herdmates as reported in an earlier study.⁸ Therefore, some calves born to MAP-infected dams in the present study might have been infected congenitally. If present, the effect of in utero transmission would have been to bias the association between ingestion of MAP DNA–positive colostrum and the risk of transmission of MAP in the calves studied in either direction (ie, toward or away from the null hypothesis). Although the present study found no association between exposure status and risk for testing positive for MAP in adulthood, potential misclassification error with respect to exposure status of the calves could not be verified given the lack of knowledge of dam MAP status. In addition, the role of other potential MAP transmission routes, which could have also been associated with a positive MAP status for dams of the study cohorts, could not be verified.

A third potential limitation of the present study involved questions regarding the accuracy of the colostrum PCR assay, serum ELISA, and culture of feces used to classify colostrum samples as MAP DNA positive or negative and to classify study animals as MAP infected or uninfected. The nested PCR assay used in the present study was found in a previous experimental validation study²¹ to have an analytic sensitivity that improves with increasing concentration of MAP (range, 1 × 10⁷ to 1 × 10⁹ cells/mL at 100%; 1 × 10³ to 1 × 10⁶ cells/mL at 75%; and 1 × 10¹ to 1 × 10² cells/mL at 50%) in colostrum and a specificity of approximately 0.89 (false-positive rate, approx 0.11). Given the previous experimental evidence suggestive of an imperfect sensitivity and specificity of the nested PCR assay, it is likely that the MAP status of some colostrum samples tested in this study was nondifferentially misclassified (ie, because of false-positive or false-negative errors or both). The effect of this error if present would have been to effectively drive the relative risk estimates toward a null value of 1.

Additionally, the serologic ELISA and bacterial culture of feces tests used to determine the MAP status of cows in the present study are known to have limited and imperfect sensitivity during the early phase of infection in cattle.^22,23,28,33 Thus, cows in this study that were shedding MAP at concentrations less than the limit of detection or were in the subclinical phase of the disease might have been misclassified as being MAP negative (ie, false negative). The consequence of potential misclassification of MAP infection status would have been to underestimate the incidence of subclinical MAP infection, thereby compromising the statistical power to detect a significant subclinical MAP risk difference between the groups (ie, exposed vs unexposed).

Notwithstanding the limitations, findings of the present study suggested lack of an added risk of subclinical MAP infection associated with ingesting MAP DNA–positive colostrum by calves. These findings are in stark contrast with results observed in other studies, albeit with different study designs and outcomes. For example, a clinical trial of the efficacy of feeding a commercially available plasma-derived colostrum replacer for the prevention of MAP transmission in calves revealed that calves fed raw bovine colostrum collected from their dams at birth were approximately twice as likely to be infected by MAP as calves fed the plasma-derived colostrum replacer when tested at 24, 36, and 54 months of age.¹⁷ Observational studies have also identified certain colostrum management and feeding practices as being significant risk factors for MAP transmission in MAP-endemic herds. In 1 such study,¹⁸ herds in which calves were fed colostrum collected from known MAP-infected cows were 87 times as likely to be infected with MAP as were herds in which such colostrum feeding practices were not the norm. Additionally, a Danish study¹⁹ revealed that calves fed colostrum collected from multiple sources (or cows) were more likely to test positive for MAP in adulthood, compared with herdmates fed colostrum collected from their own dams. Streeter et al⁸ reported successfully cultivating MAP from colostrum collected from cows subclinically infected with MAP. Most recently, Collins et al³⁴ reported that feeding hygienically collected colostrum from MAP ELISA–negative cows to calves within 6 hours of birth, among other management changes designed to limit the risk for MAP transmission in calves (eg, segregated maternity pens for ELISA-positive vs ELISA-negative cows), significantly reduced the apparent prevalence of ELISA-positive cows from approximately 12% to approximately 6% within 6 years of implementation. Finally, Pithua et al³⁵ reported that colostrum from MAP fecal culture–positive cows (vs fecal culture–negative cows) was twice as likely to be MAP positive and that withholding colostrum obtained from MAP-positive cows could reduce the risk of calf exposure to MAP-infected colostrum by ≥ 18%. Although the previous studies did not determine cause-and-effect relationships between ingestion of colostrum and subsequent infection with MAP, because of their cross-sectional design (with the exception of the study by Pithua et al¹⁷), collectively, these studies seem to provide some evidence in support of colostrum as a possible early vehicle by which calves become exposed to MAP in para-tuberculosis-endemic herds, underscoring the potential role of colostrum feeding programs in the epidemiology of MAP in dairy herds.

The difference in incidence of MAP infection (as judged via ELISA and bacterial culture test outcomes parallel interpretation) in exposed versus unexposed groups appeared biologically negligible (difference, ≤ 1%). However, a post hoc power analysis, although methodologically unappealing, revealed that had the true population difference in incidence of MAP infection in unexposed versus exposed groups been ≥ 18%, then the number of cows included in this analysis (exposed group, 69; unexposed group, 136) would have been sufficient to provide the present study with a statistical power ≥ 81% to allow for rejection of the null hypothesis at a significance level (α) ≤ 0.05, assuming the incidence of MAP infection in the exposed group was the same as that estimated from the present data.

Given the conflicting findings (compared with findings from other studies) and limitations of this study, caution should be used in interpreting the present results. The potential role that other MAP transmission mechanisms (eg, in utero or fecal-oral transmission mechanisms) may have played in this study, given the study design, should not be underestimated. Findings of this study were inconclusive. As such, the present results should not be used as a basis to discount the potential role that raw colostrum feeding programs might play in the transmission of MAP to susceptible calves, given the evidence from other studies^8,18,19 that suggest colostrum might be an early source of MAP in infected herds. Further research is warranted to establish whether the lack of a relationship between ingesting MAP-positive colostrum and risk of MAP infection observed in the present study is real. Future studies attempting to address a similar question should determine dam status for MAP fecal shedding (infection) a priori with the intent of adjusting for this factor (dam status for MAP) at the study design phase or during the time of analysis to avoid confounding bias caused by in utero transmission of MAP^29–32 or any other transmission modes unrelated to ingesting MAP-positive colostrum.