Control of Porcine Reproductive and Respiratory Syndrome: the challenge of identifying carrier animals

Abstract - Infection with PRRSV produces clinically normal, but chronically infected, carrier animals. This is old information, but it remains the most difficult problem to solve in the prevention and control of PRRSV. What percentage of pigs are carrier animals over time post infection? At present we have partial information in this area. Our laboratory has demonstrated the presence of infectious PRRSV 60 days post inoculation (PI) in 18 of 18 individually housed pigs. Among gang-housed pigs, virus was detected in 11 of 12 animals on day 77 PI. Previously, Wills et al. (1997) reported isolation of virus from 3 of 4 experimentally infected pigs on day 84 PI, with one pig remaining virus isolation positive up to day 157. Collectively, these data suggest that 100% of pigs are infected for at least 60 days PI. Some time after 60 days PI, animals begin to clear the infection. Identification of carriers is pivotal for the successful implementation of prevention and control procedures in commercial swine herds. Do we have the diagnostic technology to effectively identify carrier animals? At a minimum, developing a process to identify carriers involves: (1) determining which ante mortem diagnostic sample is most likely to contain virus and (2) evaluating the performance characteristics of diagnostic assays for the preferred ante mortem sample. Without good data on diagnostic parameters, it is not possible to confidently interpret results in the diagnostic setting. For PRRSV, tonsil and lymph nodes appear to be the tissues in which virus persists longest. Lymph nodes are not practical to collect ante mortem, but it has been shown to be relatively easy to isolate virus from tonsil scrapings collected from live animals (Wills et al., Vet. Microbiol., 1997). Presumably, more sensitive procedures (i.e., PCR) would be even more efficient in detecting virus. However, it is not certain that current PCR procedures are sufficient to achieve the level of diagnostic sensitivity required to identify carriers. For example, Wagstrom et al. (J. Vet. Diagn. Invest., 1999) assessed the diagnostic performance of a commercial RT-nPCR assay for PRRSV. The study used 297 samples of known PRRSV infection status, including serum samples from 195 individual uninfected swine to estimate diagnostic specificity and day 7 post inoculation serum samples from 102 experimentally infected animals to estimate diagnostic sensitivity. Results showed the diagnostic sensitivity of the assay to be 68.8% and diagnostic specificity to be 99.4%. Because PCR is generally perceived to be extremely sensitive, the diagnostic sensitivity of the assay was unexpectedly low. A search of the literature, however, found no evidence to suggest the results were unrepresentative of PCR performance. Overall, the data suggest that, at the present stage of development, we are not able to identify carrier animals in an accurate and cost effective manner. Future work needs to focus on further characterizing the carrier state, including the factors involved in clearing virus, and on optimizing diagnostic procedures for detecting carriers.