Evaluation of long chain lipopolysaccharides (LC-LPS) of Actinobacillus pleuropneumoniae serotype 5 for the serodiagnosis of swine pleuropneumonia.

Long chain lipopolysaccharides (LC-LPS) of Actinobacillus pleuropneumoniae serotype 5 have been evaluated and compared with a crude boiled extract (CBE) in ELISA for the serodiagnosis of swine pleuropneumonia caused by this serotype. The mean optical density (OD) obtained with the LC-LPS in ELISA using sera from negative herds as well as from animals experimentally and naturally exposed to A. pleuropneumoniae serotype 5 was not significantly different from that obtained with the CBE. However, sera from animals exposed to serotypes of A. pleuropneumoniae other than serotype 5 presented a significantly lower mean OD (P < 0.05) when the LC-LPS was used. As a consequence, it was demonstrated that a high percentage of non-specific cross-reactions were eliminated, without losing specificity. The specificity and the sensitivity of the LC-LPS- and CBE-ELISA were evaluated using two different cut-off values (the OD plus two and three standard deviations) (SD) obtained from 593 sera from negative herds. The LC-LPS appeared a more suitable antigen than the CBE, since the sensitivity and the specificity (obtained with both thresholds) were statistically improved (P < 0.01). A threshold of 0.244 (mean OD plus three SD) for the LC-LPS-ELISA seemed more suitable, since a sensitivity of 79% and a specificity of 97% was achieved. Nevertheless, it may be advisable to keep a buffer range (OD between 0.194 and 0.243) and to consider sera presenting values within this range as suspicious. In the present study, the complement fixation test presented a high specificity (97%) and a very low sensitivity (47%). A herd with animals presenting ELISA positive and CFT negative results in serology, along with the absence of suggestive lesions should not be considered as a non-infected herd.