All pigs displayed lateral recumbency or labored breathing. At 144 postinfection, all pigs including three control pigs were sacrificed by lethal injection, being given an overdose of pentobarbitone by intracardiac injection after anesthesia. The experiments were terminated on day 6 after challenge and a necropsy was performed on all pigs. In the challenge group, severe fibrinous polyserositis,

arthritis and meningitis were observed at necropsy. The results of detection of H. parasuis by bacterial isolation, nested PCR and LAMP in different samples are shown in Table 2. The LAMP from 42 samples gave a total of 23 (55%) positive results, the same result compared with nested PCR. However, LAMP gave one more positive result from brain tissue than did the culture method. On the other hand, in the control group, three selleck pigs remained clinically normal throughout the experiment and did not have lesions at necropsy. Samples obtained from the

control group were inspected using bacterial isolation, nested PCR and LAMP. All the samples were negative for H. parasuis using the three methods. Diagnosis of H. parasuis infection has traditionally been based on clinical signs, presence of lesions at necropsy and bacteriologic culture (Vahle et al., 1995). Haemophilus parasuis is a slow-growing, delicate and fastidious organism with specific nutritional requirements (Oliveira & Pijoan, 2004). Therefore, the method of identification using

culture is not always optimal, and PCR-based methods are an attractive alternative (Oliveira et al., 2001). Angen and his colleagues Selleck BYL719 PIK3C2G developed an improved species-specific PCR test for detection and identification of H. parasuis. The target sequence in 16S rRNA gene was 100% specific for H. parasuis and did not exist in species closely related to H. parasuis (Angen et al., 2007). Based on this high specificity sequence, we designed four primers for LAMP assay. In our specificity test, we also found that the target region in 16S rRNA gene did not exist in the A. pleuropneumoniae, P. multocida, Bordetella bronchiseptica, M. hyopneumoniae and S. suis, which are the common pathogens in pig respiratory problems. In the laboratory test, we found that the LAMP assay was more sensitive than nested PCR. When LAMP, nested PCR and bacterial isolation methods were used separately to test the lung tissue samples obtained from 122 pigs with an apparent infection of the respiratory tract, we found that all the samples that were positive by bacterial isolation were also positive by nested PCR and LAMP assay. Moreover, the LAMP assay demonstrated higher sensitivity, picking up 16 additional positive samples with low levels of bacteria that were missed by nested PCR (P=0.02). We also sampled lung tissue from 55 healthy pigs. All these samples were inspected by the three methods. The results showed that all samples were H. parasuis negative in the three methods.