Development and Evaluation of a Herd Health Monitoring System for Swine Operations
by G. Regula, R.M. Weigel, C.A. Lichtensteiger, N.E. Mateus-Pinilla, G. Scherba, and G.Y. Miller
In today's intensive pork production systems, herd health is
a major factor determining the profitability of the business. Among the economically
most important diseases in the Midwest are porcine reproductive and respiratory
syndrome, pseudorabies, transmissible gastroenteritis, swine influenza and pneumonia
caused by Mycoplasma hyopneumoniae and Actinobacillus pleuropneumoniae.
Several studies have evaluated the financial impact of acute outbreaks of these
diseases in swine operations (Miller & Kliebenstein, 1985, Mullan et a. 1994,
Poulson et al., 1993, Rougoor et al., 1996). However, the predominant problem
in many swine operations is a subclinical infection of the herd rather than
acute outbreaks of diseases. Infected pigs are not obviously sick or die, but
perform poorly. The economic consequences of subclinical infection can be slow
growth, decreased feed efficiency, low fertility, increased numbers of abortions
and stillbirths, and smaller litter sizes (Baysinger et al., 1997). For farms
operating on an 'all in, all out' basis, the difference in weight gain between
infected and non-infected pigs can be an additional problem, because finishing
pigs do not reach market weight at the same time.
The objective of our study was to develop a herd health monitoring system to
detect subclinical disease at an early stage. The ability of this system to
detect changes in herd health status was evaluated. We also determined the magnitude
of the impact of infection with different pathogens on weight gain, carcass
quality and reproductive performance. On the basis of this information, the
costs and benefits of specific intervention measures (e.g. vaccination, test-and-removal,
antibiotic treatment) can be evaluated. This will assist farm managers in decisions
on disease management.
Materials and Methods
In 1996, the first year of this two year study, three university farms were
monitored. Farm 1 was a pasture operation. On farm 2, the sows and nursery pigs
were kept indoors and the grower and finisher pigs outdoors. Farm 3 was a complete
confinement operation. Each farm was visited four times, with a 6-8 week interval
between visits. On every visit blood samples
were obtained from 90 pigs of different age groups. On these samples
serological testing was performed to detect antibodies to porcine respiratory
and reproductive syndrome virus (PRRSV), transmissible gastroenteritis virus
(TGEV), swine influenza virus (SIV) and Actinobacillus pleuropneumoniae
(APP). These infectious agents were chosen for monitoring because of their potential
economic impact and at least one of the farms had a recent diagnosis of infection
with these pathogens. Each farm had two cohorts of 30 pigs followed from nursery
to slaughter. Individual pigs in the cohort were identified with ear tags. At
each visit, pigs were weighed to monitor weight gain. Serum samples were collected
to monitor changes in antibody titers. Figure 1 illustrates the sampling scheme
on each farm for the first project year.
At slaughter, carcasses were examined for gross pathology: the skin was evaluated
for mange, the snout for atrophic rhinitis, the lungs for pneumonia and the
liver for parasite infestation. In order to evaluate carcass quality, back fat
and loin eye area were measured and color, firmness and marbling of the loin
eye were scored. To determine the association of subclinical infection with
reproductive performance of sows, the farrowing interval, number of liveborn,
number of stillborn and weaning weights were recorded. The influence of seropositivity
on weight gain in nursery to finisher pigs and on productivity in sows was evaluated
using a multiple linear regression analysis.
In all of the three university herds PRRSV was endemic in the
sow herd with seroprevalence rates of 11.7%, 6.7% and 51.7% respectively. TGEV
infection was prevalent in all herds throughout the year; 18.3% to 56.7% of
the sows were seropositive. However, grower and finisher pigs were seropositive
for TGEV only sporadically. On all three farms seroprevalence rates were the
highest for SIV, with averages of 63.9%, 59.7% and 63.3% respectively. Only
one farm had APP seropositive animals during a limited period of time. The surveillance
system was effective in detecting changes in disease prevalence $15%
Figure 2 shows the prevalence of antibodies against PRRSV, SIV, TGEV and APP
on each farm. Figure 3 indicates the prevalence for each age group across farms.
A multiple linear regression model was used to evaluate the influence of seropositivity
on weight gain. The outcome variable was average daily weight gain (in grams)
between the first farm visit (6-8 weeks of age) and the last farm visit (22-24
weeks of age). The mean average daily weight gain ( "
standard deviation) on farm one was 820 ("
97) grams per day, on farm two 651 ("
65) grams per day and on farm three 738 ( "
75) grams per day. For the regression model farm and cohort effects were coded
by orthogonal contrast coding. On each farm the two cohorts were compared among
each other. The pasture farm was contrasted to the other two farms and the farm
that raised finishers outdoors was compared to the complete confinement operation.
Other variables included in the original model were age at first bleeding and
weighing, sex, positive antibody titers for each of the four diseases at each
of the three bleedings and gross pathology at slaughter. A significant influence
on weight gain was found for farm and cohort effects and age at first bleeding
and weighing. Sex was not significant at an a
level of 0.05 (p=0.085), but was forced into the model as females have been
shown to have a slower weight gain if no split sex feeding is practiced. Nursery
pigs with positive titers for SIV at 6-8 weeks of age gained an average of 27
grams per day more than pigs without maternal antibodies at that age. Antibodies
for PRRSV in the nursery increased the weight gain by 73 grams per day. The
six pigs that seroconverted for PRRSV in the finisher stage had a significantly
lower weight gain (-46 grams per day) compared to the negative pigs. Of the
findings at slaughter only lung pathology influenced average daily weight gain
significantly. For each percent increase in lung tissue with pneumonia, weight
gain was decreased by 5 grams per day. Pigs that showed fibrous scars in the
lung tissue gained 23 grams per day less than pigs without visible signs of
having had pneumonia at an earlier age. In table 1 the variables that remained
in the final multiple regression model and their associated ß,
t, p and sr2 values are listed.
To evaluate the impact of subclinical disease on sow productivity, multiple
linear regression models were performed with the farrowing interval, number
of liveborn piglets, number of stillborn (including mummified fetuses) and weaning
weight as the outcome variables. The weaning weight and the number of liveborn
piglets in a litter were not associated with any of the monitored diseases.
The mean ("
standard deviation) farrowing interval was 149.1 ("
4.7) days on farm one, 146.0 days ("
22.3) on farm two and 135.8 ("
12.7) days on farm three. The farrowing interval was increased by 7.1 days in
APP positive sows (p=0.023). Sows vaccinated and seropositive for TGE had a
25.9 days shorter farrowing interval (p=0.006). Vaccinated TGE positive sows
also had a smaller number of stillborn or mummified piglets in the litter. The
average number of dead pigs in the litter was 2.7 on farm one, 0.9 on farm two
and 1.8 on farm three. A positive vaccination titer for TGE decreased this number
by 0.6 (p=0.045). The number of stillborn piglets in sows seropositive for SIV
was increased by 0.9 (p=0.018). None of the sows were vaccinated for SIV. Seropositivity
for PRRSV did not influence any of these measures of reproductive performance.
The herd health monitoring system was effective in detecting changes
in herd health status on the three farms monitored during the first year of
the study. In the multiple linear regression model for average daily weight
gain of nursery to finisher pigs antibody titers were responsible for only a
small part of the variation in weight gain. Most of the variation in weight
gain could be attributed to farm and cohort effects. Of the total variation
in weight gain, 1.41% could be explained by antibody titers at 6-8 weeks and
0.61% by seroconversion for PRRSV in the finisher phase. This could mean that
subclinical disease has only a small impact on performance compared to feeding,
genetics and management factors. However, the farms chosen for this study represented
a broad range of management types from a pasture operation to complete confinement.
Therefore the high variation between farms is expected. Antibodies for SIV or
PRRSV at 6-8 weeks of age had a positive effect on average daily weight gain.
This indicates that the titers measured are from maternal antibodies rather
than from infection. Most of these seropositive pigs were seronegative at the
second bleeding at 16 weeks of age. There was also a positive correlation between
antibody titer and weight in 6-8 week old nursery pigs.
Of the subclinical
diseases monitored, only antibodies for PRRSV had a significant impact on average
daily weight gain. However, in the groups monitored, the prevalence of antibodies
for TGEV and APP were too low to detect an impact of these diseases on weight
gain. More than 80% of the finisher pigs seroconverted for SIV. Thus a moderate
decrease in weight gain due to this infection may have remained undetected because
the sample size of seronegative pigs was too small. The second year of this
study includes a greater number of farms as well as a slightly increased cohort
size. Therefore the evaluation of the magnitude of the impact of TGEV and SIV
on weight gain will be facilitated. The prevalence of APP was low on all farms,
it might not be an important pathogen for subclinical infection in the area
we were doing our research in.
Fertility in sows was significantly decreased by several of the
monitored diseases. While litter size and weaning weights were not affected,
the number of stillborn and mummified piglets as well as the length of the farrowing
interval were influenced by the health status of the sows. Positive titers in
unvaccinated sows increased the number of stillborn piglets and increased the
farrowing interval. Positive titers from vaccination, on the other hand, improved
reproductive performance. It was surprising that positive titers for PRRSV did
not have an effect on fertility in the sows monitored in this study. Sows can
remain seropositive for PRRSV for a long time after infection. The data that
we evaluated included only the last farrowing before our visit to the farm.
Therefore an earlier infection with persistent antibodies might not have an
impact on reproductive performance during our measuring interval. However, Baysinger
et al. (1997) observed a decrease in reproductive performance that lasted for
one year after infection with PRRSV. For the continuation of the project in
1997, the sows are identified individually and blood samples are collected twice
in a 6 month interval. Thus, we will be able to evaluate possible changes in
serological status over time.
Baysinger, A.K., Dewey, C.E., Straw, B.E., Brumm, M.C., Schmitz,
J., Doster, A., Kelling, C. 1997. Sw.HealthProd. 5(5):179-187.
Miller, G.Y., Kliebenstein, J.B. 1985. The economic
impact of clinical transmissible gasteroenteritis for swine producers participating
in the Missouri mail-in-record program. Prev.Vet.Med. 3:475-488
Mullan, B.P., Davies, G.T., Cutler, R.S. 1994. Simulation of the
economic impact of transmissible gastroenteritis on commercial pig production
in Australia. Aust.Vet.J. 71(5):151-154
D., Marsh, W., Morrison, R., Dial, G. 1993. A methology for evaluating the financial
consequences of a disease outbreak of transmissible gasteroenteritis and pseudorabies
virus. Prev.Vet.Med. 16:61-63
C.W., Dijkhuizen, A.A., Huirne, R.B.M., Marsh, W.E. 1996. Impact of different
approaches to calculate the economics of disease in pig farming. Prev.Vet.Med.