Composite testing for ante-mortem diagnosis of Johne's disease in farmed New Zealand deer: correlations between bacteriological culture, histopathology, serological reactivity and faecal shedding as determined by quantitative PCR

Abstract

Background: In the absence of overt clinical signs of Johne's Disease (JD), laboratory based tests have largely been limited to organism detection via faecal culture or PCR and serological tests for antibody reactivity. In this study we describe the application of quantitative faecal PCR for the detection of Mycobacterium avium subsp. paratuberculosis (MAP) in New Zealand farmed deer to quantify the bacterial load in cervine faecal samples as an adjunct to an existing serodiagnostic test (Paralisa™) tailored for JD diagnosis in deer. As ELISA has potential as a cheap, high throughput screening test for JD, an attempt was made to assess the sensitivity, specificity and positive/negative predictive (PPV/NPV) values of Paralisa™ for estimating levels of faecal shedding of MAP as a basis for JD management in deer.

Results: Correlations were made between diagnostic tests (ELISA, qPCR, culture and histopathology) to establish the precision and predictive values of individual tests. The findings from this study suggest there is strong correlation between bacterial shedding, as determined by faecal qPCR, with both culture (r = 0.9325) and histopathological lesion severity scoring (r = 0.7345). Correlation between faecal shedding and ELISA reactivity in deer was weaker with values of r = 0.4325 and r = 0.4006 for Johnin and Protoplasmic antigens, respectively. At an ELISA Unit (EU) cutoff of >50 (Johnin antigen) the PPV of Paralisa™ for significant faecal shedding in deer (>104 organisms/g) was moderate (0.55) while the NPV was higher (0.89). At an EU cutoff of ≥ 150, the PPV for shedding >105 organisms/g rose to 0.88, with a corresponding NPV of 0.85.

Conclusions: The evidence available from this study suggests that Paralisa™ used at a cutoff of 50EU could be used to screen deer herds for MAP infection with sequential qPCR testing used to cull all Paralisa™ positive animals that exhibit significant MAP faecal shedding.

Figure 1

Correlation between MAP titre as determined by qPCR and faecal culture on HEYM. Correlation of faecal qPCR data against HEYM solid culture based on cumulative NVSL JD Proficiency Panel Samples for 2008, 2009 and 2010; n = 72 (of a total of 78 individual faecal samples, two did not have associated colony counts and four were disregarded for official grading purposes). Spearman Correlation = 0.9325; p <0.0001.

Figure 2

Correlation between MAP titre as determined by qPCR and Histopathological Lesion Severity Score. Correlation of faecal MAP titre as determined by qPCR and Histopathological Lesion Severity Score; n = 40, Spearman Correlation r = −0.7345; p <0.0001.

Figure 3

Correlation between ELISA reactivity and MAP titre as determined by qPCR. Panel A; Correlation of faecal MAP titre as enumerated by qPCR vs ELISA (Paralisa™) reactivity of matched cervine blood samples (Johnin antigen); Spearman correlation r = 0.4325; p <0.0001. Panel B; Correlation of faecal MAP titre as enumerated by qPCR vs ELISA (Paralisa™) reactivity of matched blood samples (Protoplasmic antigen); Spearman correlation r = 0.4006; p < 0.0001. n = 663. Red circle = Overall Paralisa™ result = positive; Light green circle = Overall Paralisa™ result = negative.