Johne's disease (JD) (paratuberculosis) is a chronic intestinal infection of cattle, sheep, goats, camelids, and wild ruminants caused by Mycobacterium avium subsp. paratuberculosis (MAP). In cattle, the clinical disease is characterized by diarrhea, weight loss, and edema associated with protein-losing enteropathy. Infected animals can shed MAP organisms in their feces for months to years before they show clinical signs. As a result, these undetected shedders are responsible for spread of the disease before they themselves are suspected as being infected. The most recent studies looking at the prevalence of MAP infection in cattle in the United States suggest that nearly 70% of all US dairy herds, and 95% of herds with more than 500 cows, are infected.
In contrast, the prevalence of JD in alpacas in the United States is unknown. There are individual case reports[3, 4] but no surveys of regional or nationwide prevalence. In 1994, llamas were banned from use as pack animals in national parks in the southwestern United States, for fear of transmission of MAP to the native bighorn sheep, yet there were no data to support this policy.[5, 6]
To date, the diagnosis of MAP infection in alpacas has been problematic. Like cattle, alpacas infected with MAP develop a granulomatous lesion in the small intestines, leading to protein-losing enteropathy and severe weight loss.[4, 7] However, alpacas with JD often do not develop the intractable diarrhea that accompanies the weight loss in cattle, or might develop it terminally, so owners and veterinarians might not be alerted to the possibility of a MAP infection. In addition, the weight loss is often obscured by the alpacas’ fleece, so that JD will not be suspected on the basis of casual observation alone. Diagnosis in the past has also been complicated by the lack of diagnostic tests with appropriate sensitivity to detect MAP infection in camelids. Serum tests (ELISA) for antibodies to MAP have been adapted for use in camelids. However, the sensitivity of these tests to detect MAP infection is between 50 and 70%, which is most likely explained by the fact that as an intracellular pathogen, MAP might not induce a strong humoral immune response until late in the infection.[9, 10] Similar to cattle, tests relying on the detection of antibodies could be more useful at the herd level, and not for individual animal diagnosis. Bacteriologic culture of MAP from feces, often considered the “gold standard” for diagnosis, can require 8 to 16 weeks before colonies can be identified. Recent advances in diagnostic technology have led to the application of the real-time polymerase chain reaction (RT-PCR) to quickly and accurately identify MAP in fecal specimens. In bovine fecal specimens, the sensitivity and specificity of the test were estimated to be around 60 and 97%, respectively. However, to the authors’ knowledge, the limits of MAP detection in alpaca feces by use of PCR have not been determined.
The authors hypothesized that the combination of inapparent or unrecognized clinical signs, as well as difficult laboratory confirmation, might have led to an underreporting of MAP infection in alpacas in the United States. The objectives of this study were to (1) validate the PCR test for use in alpaca feces, and define the limits of MAP detection in these samples; and (2) estimate the prevalence of MAP shedding in the feces of alpacas presented to 4 veterinary teaching hospitals in the United States.
- Top of page
- Materials and Methods
The lowest MAP dilution detectable via PCR was 243 MAP CFU/g of feces (Table 1). At that concentration, MAP growth was also detectable on 3 of the 4 HEYM tubes with 1 MAP CFU/tube. The lowest dilution of MAP to trigger a TTD in the commercial detection system4 culture medium was 1,216 CFU/g of feces. The Spearman rank correlation coefficient between the PCR Ct value and the number of CFU/g of feces on HEYM was −1.0 (P = .0004). The Spearman rank correlation coefficient between the number of CFU/g and the TTD in MGIT was −0.972 (P = .003). Finally, the Spearman rank correlation coefficient between Ct and TTD was 0.991 (P < .0001).
Table 1. Real-time polymerase chain reaction (RT-PCR) and bacterial culture results for 10 dilutions of a test strain of Mycobacterium avium subsp. paratuberculosis (MAP) added to negative alpaca feces
|MAP Concentration (CFU/g feces)||RT-PCR Triplicates (Ct)||HEYM Culture on 4 Tubes (MAP CFU/tube)||Liquid Culture TTD (days)|
|1.9 × 107||20,21,26||>300,>300,>300,>300||7.7|
|3.8 × 106||21,22,22||250–300,100–150,250–300,150–200||10.4|
|7.6 × 105||25,25,25||250–300,>300,250–300,>300||14.6|
|1.52 × 105||28,28,29||150–200,150–200,100–150,150–200||16.9|
|3.04 × 104||30,30,31||12,12,6,19||22.3|
A total of 180 fecal samples were obtained from the 4 participating referral hospitals (Cornell, 31; Oregon, 61; Tufts, 50; and UPenn, 38). Of the 180 fecal samples, 10 (6%) were positive on PCR with Ct values ranging from 33.0 to 39.6 cycles. The 95% confidence interval was determined to be between 3 and 9% of the population. None of the samples was positive on HEYM culture. The prevalence of PCR-positive animals was variable between referral hospitals, with 0, 1, 2, and 7 test-positive animals for Cornell, UPenn, Oregon, and Tufts, respectively; or 0, 3, 3, and 14% of the number of tested animals for Cornell, UPenn, Oregon, and Tufts, respectively. Three of the 7 PCR-positive alpacas from Tufts were from the same farm. The 2 PCR-positive alpacas from Oregon came from 2 different farms. The 10 PCR-positive alpacas (8 females, 2 males) ranged from 1 to 10 years of age (median, 6 years). Their body condition scores ranged from 2/10 to 9/10 (median, 5/10). The presenting complaints varied between animals and were not limited to the gastrointestinal system. Those included weight loss/diarrhea (2), mandibular swelling (1), dystocia (1), corneal ulcer (1), colic (2), and recumbency (1). The 2 remaining animals were healthy on presentation and included a dam admitted with a sick cria, and a companion admitted with the alpaca with the corneal ulcer. Of the 10 PCR-positive animals, 4 had diarrhea on presentation. Final diagnoses were available for 6 of the 8 sick alpacas, and included copper deficiency (1), tooth root abscess (1), uterine tear (1), keratopathy (1), abdominal abscess (1), and rickets (1). The 2 remaining alpacas with “unknown” diagnoses included 1 alpaca admitted for recumbency, and for which a final diagnosis could not be determined; and 1 alpaca admitted for colic and diarrhea, which resolved shortly after its admission to the hospital.
- Top of page
- Materials and Methods
The 1st objective of this study was to validate the PCR test for use in alpaca feces, and define the limits of MAP detection in these samples. The main advantage of RT-PCR over traditional MAP culture is that it provides a faster result regarding MAP fecal shedding of a given animal. In a previous study, RT-PCR results were highly and inversely correlated with HEYM culture results, where the higher the number of MAP CFUs per sample, the lower the Ct value for that sample, with the advantage that results are available in days instead of months. This study confirms the utility of RT-PCR for estimating the level of MAP shedding in alpaca feces. It is recognized that MAP organisms can clump in culture suspension or in feces, so that CFU counts might not correspond 1 : 1 with the number of organisms in the sample. However, our results showed that, as the number of organisms or clumps of organisms (CFU) are reduced in the sample through dilution, results of HEYM culture (CFU counts), liquid culture (TTD), and RT-PCR (Ct values) all changed in a proportional and highly correlated fashion. Thus, the CFU values shown in this study should not be construed to indicate the specific number of organisms in the samples because of the possibility of clumping, but it is reasonable to assume that the 2 values are related, such that higher CFU values represent a greater number of organisms in the sample. Finally, many clinicians have experience interpreting HEYM CFU values from fecal samples, so that these values represent a convenient and familiar reference to which RT-PCR Ct values can be compared.
The 2nd objective of this study was to estimate the prevalence of MAP shedding in feces of alpacas presented to 4 veterinary teaching hospitals in the United States. In the study reported here, the overall prevalence for the population tested was 6%, with a 95% confidence interval between 3 and 9%. The population tested was comprised of alpacas of all ages admitted to 4 veterinary teaching hospitals including 3 teaching hospitals located in the Northeast and 1 from the northwestern part of the United States. It is also acknowledged that the prevalence reported here was obtained from a population comprised of hospitalized alpacas, and might not be a reflection of the prevalence of MAP fecal shedding in the general alpaca population in the United States. Regardless, a MAP fecal shedding prevalence of 6%, even in hospitalized patients, represents a higher prevalence than would be suggested by cases of JD in alpacas seen at those hospitals (no confirmed cases) during the study period. Possible explanations for the discrepancy between the number of clinical cases of JD in alpacas admitted to the participating veterinary hospitals and the prevalence of MAP fecal shedding in the population tested might include 1) some animals with clinical JD that are misdiagnosed; 2) the minority of infected camelids that go on to develop clinical JD; 3) some animals that are only transiently infected with MAP and self-cure the infection; or 4) some animals that are passive shedders (pass-through) and are not really infected with MAP.
Of the 7 PCR-positive alpacas from 1 veterinary teaching hospital, 3 were from the same farm. This fits with previous reports of both farm-related outbreaks and sporadic individual cases.[3, 4] Genotyping of the MAP isolates could not be performed because MAP was not cultured. Therefore, it is not possible to know if the farm with multiple positive animals harbors a unique MAP genotype or several different genotypes. In a case report describing high MAP fecal shedding in an alpaca and its implications for the rest of the herd, several MAP genotypes were found in the described herd. In the same report, 1 positive alpaca was found to be infected with 2 different MAP genotypes. In the study reported here, it was also not possible to know if the MAP strain(s) were of bovine, ovine, or other origins. It is believed that most strains can infect across ruminant species lines.[7, 15] The practice of feeding bovine colostrum to neonatal crias could represent a possible source of MAP introduction within a herd. Cohousing with cattle or grazing of pastures fertilized with cattle manure could represent additional routes of transmission. Interestingly, in the study presented here, the farm with the 3 positive alpacas also housed cattle. Further epidemiologic studies are needed to determine risk factors associated with the presence of MAP on alpaca farms.
Of the 10 PCR-positive animals, 4 displayed potential signs of JD (weight loss and/or diarrhea) on admission. However, 3 of those 4 alpacas were diagnosed with other diseases, copper deficiency (1), rickets (1), and abdominal abscess (1), which could also explain their clinical signs. The remaining alpaca with diarrhea was in excellent body condition and was presented for colic that quickly resolved, although the cause of her colic episode was never determined. Because these animals were discharged alive from the hospitals, postmortem examinations were not performed. Follow-up information regarding response to treatment after discharge was not available. It is therefore impossible to know for certain if JD played a role in the clinical presentation of those animals or if the MAP fecal shedding was purely an incidental finding. This emphasizes the importance of ruling out other known causes of diarrhea or weight loss or both when presented with an alpaca displaying chronic weight loss, change in fecal character, /or hypoproteinemia.
The interpretation of a positive PCR result in an asymptomatic alpaca can represent a challenge for a clinician. The RT-PCR for MAP is highly specific (around 97%) in other species. In this study, most PCR-positive prevalence samples had high Ct values, were positive in only 1 well, and all were negative on HEYM culture. All this suggests low levels of fecal shedding. Although in cattle, it is not uncommon that fecal samples with Ct values near the positive cut-off (38–42) are negative on culture; it is acknowledged that, based on the results obtained in the validation part of the study presented here, PCR-positive animals with a Ct value between 35 and 38 were expected to have at least 1 MAP CFU per HEYM tube. The difference in culture sensitivity between the 2 experiments presented here could be attributed to several factors: 1) fecal samples from the prevalence animals were shipped and frozen for up to 8 months before being cultured, which might have affected MAP viability; 2) the PCR-positive animals from the prevalence study might have been harboring a different strain of MAP that can be more difficult to culture than the test strain used in the validation part of the study; 3) the test strain used in the validation part of the study had already gone through one passage and therefore might have been more adapted to laboratory conditions; and 4) in animals that are naturally infected, MAP might be more intracellular and can be more difficult to grow than in feces that have been spiked with a known amount of MAP.
In cattle and other ruminant species including camelids, animals that are shedding high numbers of MAP in their feces can contribute to substantial environmental contamination,[1, 4] and the presence of 1 or several high shedders can contribute to “passive shedding” or “pass-through shedding” of MAP by herd mates. As MAP is deposited on pasture by infected animals, it can be ingested by other animals and detected in their feces by RT-PCR.1 Those “passive shedders” typically have a high Ct number and are found negative on follow-up fecal RT-PCR testing, especially after removal of a high shedder from a group. In the study reported here, alpacas were tested only once. Whether these low-level shedders were transiently infected, colonized but not likely to develop JD, or simply identified before the development of JD remains unknown. Furthermore, as documented in other species, an animal can shed significantly different levels of MAP over relatively short (days) and long (weeks to months) periods of time. Clinicians and owners must keep in mind that intermittent shedding could potentially lead to the misclassification (false negative) of animals, especially when tested only once. Additional studies aimed at following the progression of PCR-positive alpacas are necessary to evaluate the clinical significance of a positive result.
In conclusion, the authors acknowledge that this is a limited study, which included a small number of animals from 4 teaching hospitals, and involving a disease process that is still poorly characterized in camelids. However, from this experiment, we were able to show that RT-PCR could be used to identify MAP fecal shedding in alpacas; and that MAP fecal shedding was found in 6% of the alpacas presenting to the 4 veterinary teaching hospitals included in this study.