Evaluation of Italian Aujeszky's disease control programme in two regions of northern Italy in 1997

Abstract - On 1 April 1997, a National Control Programme for Aujeszky's Disease (AD) was made compulsory by the Italian Ministry of Health. The plan is based on the use of minimal sanitary and hygienic rules, vaccination of all pigs with the inactivated gE-deleted vaccine, completion of a questionnaire on all farms to collect individual epidemiological data, and annual sampling of all breeding herds, in order to detect a minimum prevalence of 0.80 with a 95% CI. Lombardy (L) and Emilia Romagna (ER) are two intensive pig-producing regions of northern Italy, in which the progress of the programme is being monitored closely. Questionnaires were completed and blood samples were collected in both breeding and mixed breeding-fattening herds. In L, all 1014 existing farms (a total of 8938 pigs) were both interviewed and sampled, whereas in ER only 160 (a total of 1343 pigs) of the 822 interviewed farms were also sampled (19.46%). Such sampling included all larger herds. The minimum sample size was 3 animals per category (primiparous and multiparous sows, growers and fatteners, where applicable). Both univariate ( and Fisher's exact tests) and multivariate (logistic regression) analyses were carried out, in order to detect any significant risk factor for infection with ADV. Most farms are located in flat country, especially in L (96.4% of all tested farms), less so in ER (70.2%). Nearly all are intensive breeding farms, with 99.8% of them keeping animals in totally indoor premises. Production cycles are either open or closed and less frequently mixed (4.3% of L farms, none of ER). On the average, herd size is larger for L farms: nearly 68% of L herds have more than 100 breeders, as opposed to 32% for ER. Replacement gilts are either produced internally or purchased from a different farm; for 65% of ER farms internal replacement is the rule (vs. 45.4% for L), which is a reflection of the higher proportion of closed-cycle herds in this region (57.5% vs. 46.1%). The data collected during the first year of the campaign showed an overall prevalence of ADV of 0.81 for L and 0.82 for ER . A number of variables were found to be significantly associated with seroprevalence at the univariate level. Variables showing a p-value < 0.20 at the univariate level and having a reasonably low number of missing values were further used to build a multivariate logistic regression model. Several models were tested and non-significant variables were eliminated step by step, until the final model was selected for each region. The model for Lombardy indicates that the variable CYCLE has the strongest association with the outcome variable (a seropositive finding on the farm). The odds ratios (ORs) for closed-cycle and mixed-cycle herds, compared with open-cycle herds, were respectively 4.19 and 6.41. Herd size ( > 100 breeders) was also found to be a risk factor for L pig farms , as was closeness to other piggeries, as evidenced by a higher risk of positivity for farms with more than pigs within a 6 km radius. Production and selling of breeders, on the other hand, appears to have a protective effect against AD . Similar findings were obtained for ER, the main difference being that production cycle did not show any significant association with serological status of the herd. The odds ratios for the variables N. OF BREEDERS, N. PIGS < 6 KM RADIUS, and SELLING OF BREEDERS were 4.99, 10.41, and 0.04, respectively. The sample size used (3 animals per category) did not allow us to determine within-herd prevalence. It was however possible to make some comparisons between different categories of animals. On the average, the presence of antibodies in multiparous sows was twice as high as in primipares ( and 2.26 respectively for L and ER). Such a difference was more marked for closed-cycle farms than for open-cycle farms ( and 3.22). In closed-cycle herds, a similar finding concerned fattening pigs: fatteners were twice as likely to be seropositive as growers in L and four times as likely in ER . The conclusions of this study are that seroprevalence for ADV in Northern Italy in 1997 exceeded 80%, and that infection was significantly more frequent in older animals (multiparous sows and fatteners) than in younger categories, especially in closed-cycle herds, indicating that the virus keeps spreading within the farm. It should therefore be recommended that fattening pigs receive a third vaccination, in order to avoid them acting as a reservoir for the virus.