CASE NOTES


Monitoring the prevalence and impact of enteric disease in pigs

Alison Collins, NSW DPI, EMAI

Posted Flock & Herd March 2015

INTRODUCTION

Lawsonia intracellularis is one of the most common enteric infections in grower and finisher pigs, along with Brachyspira spp. and Salmonella. Diagnosis of the cause of scouring relies on histopathology, serology, culture and polymerase chain reaction (PCR). While histopathology is the gold standard in diagnosis for many enteric diseases, only very sick pigs are necropsied, so mild clinical and sub-clinical disease are often undiagnosed. Serology has proven useful in determining the timing and prevalence of L.intracellularis and Salmonella infection (Harris, 2003; Stege, 2004), but is of less value for detecting herds positive for Brachyspira hyodysenteriae, due to specificity issues (Song et al., 2012). Antibodies to L.intracellularis are not detectable until three weeks after infection, so only indicate previous infection, but are extremely useful in optimising the timing of vaccination prior to exposure.

Conventional PCR is a very sensitive diagnostic assay and can identify pigs currently infected with all three pathogens, but positive PCR for any one of the three pathogens doesn't prove that a single pathogen is the cause of diarrhoea. In many cases, multiple pathogens are present in scouring pigs (La et al., 2006). Quantitative PCR assays have recently been developed that can calculate the numbers of L.intracellularis, Brachyspira hyodysenteriae and B.pilosicoli in pig faeces in real time (Reiner et al., 2011). L.intracellularis numbers in pig faeces correlate significantly with faecal consistency scores (R2 = 0.7), the prevalence of histopathological lesions (R2 = 0.7), the extent of gross pathology (R2 = 0.5) and negatively with average daily gain (R2 = - 0.44) (Pedersen et al., 2012; Collins et al., 2014).

Regardless of the pathogen and disease, any new diagnostic test needs to be evaluated for sensitivity and specificity relative to the gold standard assay. This paper focuses on the validation of a new qPCR assay for L.intracellularis, including the specificity and sensitivity in pooled faecal samples, a necessary requirement for reducing test costs. The variation in L.intracellularis load over time (with and without management changes) also had to be quantified in order to accurately monitor disease control over time. Lastly, the number of L.intracellularis detected in faeces by qPCR was correlated with average daily gain to test whether the qPCR would be useful for diagnosis of sub-clinical and clinical disease on a pen or herd basis. A good correlation between ADG and L.intracellularis numbers would demonstrate that the qPCR was able to evaluate the efficacy of different treatments on proliferative enteropathy (PE) control.

This paper summarises some of the applications of this new L.intracellularis qPCR to diagnose both clinical and sub-clinical proliferative enteropathy (PE) and to evaluate management strategies for disease control on commercial pig farms. In the absence of appropriate disease control, PE can cost producers as much as A$8 to A$13 per pig in lost revenue, with both clinical and sub-clinical PE causing reduced growth, variation in pig weights, poor feed conversion and increased days to slaughter.

METHODS

Pooling of faecal samples

All animal experiments were performed according to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. Initial studies were undertaken to determine the sensitivity of the L.intracellularis qPCR in pooled faecal samples. Pools of faeces were prepared to mimic populations with 10%, 20%, 40%, 60% or 100% of pigs with either sub-clinical or clinical PE. In addition, random pools of faecal samples were prepared from individual samples with known numbers of L.intracellularis to determine the optimal pool size. Individual faecal samples from pigs experimentally challenged with L.intracellularis or negative control pigs were randomly mixed into pools of 1, 5 or 10 samples per pool. DNA was extracted from 0.1g of faecal samples using a MagMax DNA extraction kit (A&B Applied Biosystems, California, USA) as previously described (Collins et al., 2011). DNA was amplified by qPCR and the mean number of L. intracellularis per gram of faeces (and standard deviation) was determined and Log10 transformed. Due to the small sample sizes, these sample distributions were used to generate results for much larger populations (50,000) using the Monte Carlo method (Metropolis and Ulam, 1949).

Variation in L.intracellularis numbers over time

Pooled faecal samples were collected from three consecutive batches of pigs from the same two commercial herds over six months to determine the normal variation in L.intracellularis numbers in the absence of management changes. Either 10 or 20 pooled faecal samples (5 pigs per pooled sample) were collected from five different age groups of pigs between 7 and 18 weeks of age on each occasion. The two subsequent samples were collected from the same age groups at about 7 week intervals. The number of L. intracellularis detected per gram of faeces was transformed (natural log) to normalise data and distribution plots were compared over time.

Critical number of L.intracellularis for clinical disease

Herds were selected where weight gains and feed intake could be calculated relatively frequently on a pen basis. The selected herds were all high health status with sows vaccinated against porcine circovirus 2 (PCV2), Pasteurella multocida, Actinobacillus pleuropneumoniae, Erysipelas, Parvovirus, Leptospirosis, Glassers disease and Mycoplasma hyopneumoniae. Pre-trial screening in each herd indicated that L.intracellularis was present in grower and finisher pigs, along with Salmonella and Brachyspira spp.

At each farm, pigs were part of other nutrition or health research trials that had the potential to impact on production and also on the animal's health. Pigs in herds 1 and 2 were separated into 4 different treatments and pigs in herd 3 were separated into 7 treatments, providing 15 different pig populations. Genetics, hygiene, disease, diet and housing conditions were different between these populations to ensure that the results from this work could be applied more widely to other herds.

At day 0, pigs were randomly selected and housed in pens of between 22 and 40 pigs per pen. Pooled pen faecal samples were collected from the three herds over the same periods that pen feed intakes and weight gains were also recorded (Table 1). Numbers of L.intracellularis per gram of faeces were calculated from standard curves plotted from faeces seeded with known numbers of Lawsonia (104 to 108 Lawsonia per gram of faeces). The Lawsonia numbers were Log10 transformed to normalize the distribution of data. A spearman's rank correlation coefficient (non-parametric) was used to test the association between L.intracellularis numbers and ADG, feed intake and feed efficiency for each individual herd (Genstat 16th edn, VSN International, 2013).

Table 1: Sampling protocol for faecal sampling and weight gains on three commercial herds

RESULTS

Pooling of faecal samples

L.intracellularis could be detected in a pool of 10 faeces containing only one clinical sample by qPCR; however the sensitivity was much lower with sub-clinically affected pigs. At least six sub-clinically affected pigs were needed in a pool of ten samples to detect L.intracellularis (Collins and Barchia, 2013a).

The number of L.intracellularis detected in individual samples versus pooled samples was analysed using 95 % confidence intervals from means and standard deviations for each pool size. The mean number of L.intracellularis in individual samples (4.94 log transformed) lay within the range of the mean number of L.intracellularis in pools of 5 samples (4.87 to 7.06), indicating that pooling 5 samples provided a representative estimate of the number of L.intracellularis (Table 2). However, pooling 10 samples provided a poor representation of L.intracellularis numbers, as the range for 10 pooled faeces (between 5.18 and 6.77) did not include the population mean for individual samples.

Table 2: The mean and standard deviation (SD) of L.intracellularis numbers detected in individual or pooled faecal samples (log transformed) and the 95% confidence intervals (Collins and Barchia 2013a)

Variation in L.intracellularis numbers over time

The variation in the number of L.intracellularis excreted in consecutive batches of pigs in herd A was minimal, in the absence of management changes. The population distribution for herd A shows that the mean was between 8 and 9 (natural log) over the three periods (Figure 1A). In the absence of management changes, the mean number of L.intracellularis remained within one standard deviation of the mean over 3 consecutive batches of pigs. However, when stocking rate or medication was altered, as occurred in herd B between batches 2 and 3, the mean and distribution of L.intracellularis shifted to the left (Figure 1B). This variation in mean L.intracellularis numbers was greater than 2 standard deviations from the mean, indicating that the L.intracellularis load varied significantly between batches of pigs in the presence of management changes (Collins and Barchia, 2013b).

Figure 1. The frequency distribution of L.intracellularis numbers over six months in three consecutive batches of pigs on a herd without significant management changes (A), and on a second herd where both stocking rate and medication were altered between batches 2 and 3 (B)

Critical number of L.intracellularis for clinical disease

In the third study L.intracellularis numbers were correlated with average daily gains over a range of commercial conditions to determine whether the qPCR could diagnose sub-clinical or clinical PE in the field using pooled faecal samples.

L.intracellularis numbers in herd 1 peaked when pigs were between 17 and 22 weeks of age, when 76% of pens were positive for L.intracellularis, and more than 106 L.intracellularis were excreted per gram of faeces in 49% of pens. L.intracellularis numbers were negatively correlated with both ADG (r = -0.23, p = 0.037) and feed intake (r = -0.34, p = 0.002) using Spearman's correlation coefficient, ie. pigs excreting higher numbers of Lawsonia were most severely affected with poor growth and reduced feed intake. L. intracellularis numbers were not significantly correlated with feed conversion efficiency (p = 0.14).

In herd 2, L.intracellularis numbers peaked between 14 and 16 weeks of age, when more than 84% of pens were positive and more than 106 L.intracellularis were excreted per gram of faeces in 40-44% of pens. Negative correlations between Log10 L.intracellularis numbers and ADG and feed intake were close to significant (r = -0.305, p= 0.146, and r = -0.266, p= 0.098 respectively) between 12 to 15 weeks of age. No significant correlations between feed conversion efficiency and Log10 Lawsonia numbers were observed.

In herd 3, L.intracellularis numbers increased between 11 and 15 weeks of age during a sub-clinical outbreak of ileitis, with excretion levels peaking between 106 and 108 L.intracellularis per gram of faeces. Negative correlations between Log10 L.intracellularis numbers and ADG were significant over this period (r = -0.350, p= 0.016). A clinical outbreak of the haemorrhagic form of proliferative enteropathy also occurred in medicated pigs in this herd 3 weeks after medication was removed. L.intracellularis numbers peaked at between 108 and 1010 per gram of faeces and significant negative correlations were observed between Log10 L.intracellularis numbers and ADG and feed intake (r = -0.573, p < 0.001 and r = -0.616, p < 0.001 respectively). A significant positive correlation was also observed between feed conversion efficiency and Log10 Lawsonia numbers (r = 0.376, p = 0.011).

DISCUSSION

This paper has demonstrated some applications of a new quantitative assay for the pathogen L.intracellularis in pig faeces, including diagnosis of PE in scouring pigs. Elevated L.intracellularis numbers correlated with scouring and also with gross and histological lesions in the ileum. In addition, large reductions in average daily gain (131 g/day) occurred when pigs excreted more than 108 L.intracellularis per gram of faeces (Collins and Barchia, 2014). It is likely that lesions and clinical signs of PE occur in pigs excreting less than 108 L.intracellularis, but we can now state with certainty that pigs excreting more than 108 L.intracellularis have lesions and clinical signs of PE. This does not preclude the possibility that other pathogens are present in scouring pigs, and possibly exacerbating disease, but L.intracellularis is definitely a significant cause of scouring and weight loss.

The quantification of L.intracellularis numbers can also be used by vets and producers to determine if infection in their herd is below the critical threshold where production losses occur. If Lawsonia numbers are above this critical threshold, vets can alter management practices in a timely fashion to avoid continued production losses.

The absence of significant variation in L.intracellularis numbers over consecutive batches of pigs without management changes indicates that the pathogen load measured by qPCR can be used to monitor PE control over time. Conversely, changes in stocking rate or in-feed medication led to significant variation in L.intracellularis numbers over time, which may impact on the growth and health of pigs. In the past, the effect of management changes could only be judged by changes in the prevalence or severity of clinical PE, including scouring and weight gains. The increased sensitivity and specificity of the qPCR relative to monitoring clinical signs suggests that the qPCR will enable producers to more accurately monitor the effect of management changes, even in the absence of clinical disease. The accurate estimation of L.intracellularis numbers from five pooled faecal samples randomly collected from the pen floor increases the cost-effectiveness of this assay to monitor PE control in the field.

Pig producers and veterinarians are under increasing pressure to reduce antibiotic use for the treatment and prevention of diseases. However, producers and veterinarians have been reticent to remove antibiotics for PE control for fear of outbreaks of the haemorrhagic form of PE. Removal of antibiotics in naive finisher pigs can lead to significant disease and mortalities (Collins et al., 2001). The L.intracellularis qPCR will allow producers to measure the level of infection prior to antibiotic removal, reducing the risk of mortalities and haemorrhagic PE.

The final application for the L.intracellularis qPCR is on-farm evaluation of different treatments for PE control. The strong negative correlation between L.intracellularis numbers and average daily gain allows producers to compare an old and new treatment for PE control on their own farm, with its associated environmental and genetic factors. Testing pooled pen faecal samples for L.intracellularis load will provide local evidence for the impact of different treatments or management practices. Initial studies in large commercial herds have evaluated the efficacy of vaccination, antibiotic medication and improved hygiene (disinfection) to control PE and reduce the associated production losses.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the technical support provided by Stephen Heavener, a graduate trainee with NSW DPI and Dr Damian Collins (NSW DPI) for substantial statistical support. Piggery staff and consulting veterinarians also provided invaluable support in collecting samples and production data. Financial support was provided by the Australian Pork CRC for High Integrity Pork.

REFERENCES

  1. Collins AM, Barchia IM. The critical threshold of Lawsonia intracellularis in pig faeces that causes reduced average daily gain in experimentally challenged pigs. Vet Microbiol 2014;168:455-8
  2. Collins AM, Barchia IM. Optimal pooling of faeces to quantify Lawsonia intracellularis in clinically and sub-clinically affected pigs. In: Manipulating Pig Production XIV, Pluske J (Ed) Australasian Pig Science Association, Werribee, Victoria, Australia, 2013a:200
  3. Collins AM, Barchia IM. Variation in the number of Lawsonia intracellularis shed in commercial pig herds over time. In: Manipulating Pig Production XIV, Pluske J (Ed) Australasian Pig Science Association, Werribee, Victoria, Australia, 2013b:199
  4. Collins AM, Fell S, Pearson H, Toribio J-A. Colonisation and shedding of Lawsonia intracellularis in experimentally inoculated rodents and in wild rodents on pig farms. Vet Microbiol 2011;150:384-388
  5. Collins AM, van Dijk M, Vu Ngoc Q, Pozo J, Love RJ. Immunity to Lawsonia intracellularis. In Proceedings of the Allen D Leman Swine Conference, Minneapolis, MN, USA, 2001:67-69
  6. Harris IT. Serologic basis for assessment of subclinical Salmonella infection in swine: Part 1. J Swine Health Prod 2003;11:247-251
  7. La T, Collins AM, Phillips ND, Oksa A, Hampson DJ. Development of a multiplex-PCR for rapid detection of the enteric pathogens Lawsonia intracellularis, Brachyspira hyodysenteriae, and Brachyspira pilosicoli in porcine faeces. Lett Appl Microbiol 2006; 42:284-288
  8. Metropolis N, Ulam S. The Monte Carlo method. J Amer Stat Assoc 1949;44:335-341
  9. Pedersen KS, Stahl M, Guedes RMC, Angen O, Nielsen JP, Jensen TK. 2012. Association between faecal load of Lawsonia intracellularis and pathological findings of proliferative enteropathy in pigs with diarrhoea. BMC Vet Res 2012; 8:198
  10. Reiner G, Hillen S, von Berg S, Kixmoller M, Willems H. Analysis of bacterial load and prevalence of mixed infections with Lawsonia intracellularis, Brachyspira hyodysenteriae and/or Brachyspira pilosicoli in German pigs with diarrhoea. Berl Munch Tierarztl Wochenschr 2011;124:236-41
  11. Song Y, Frey B, Hampson DJ. The use of ELISAs for monitoring exposure of pig herds to Brachyspira hyodysenteriae. BMC Vet Res 2012;8:6
  12. Stege H, Jensen TK, Moller K, Vestergaard K, Baekbo P, Jorsal SE. Infection dynamics of Lawsonia intracellularis in pig herds. Vet Microbiol 2004;104:197-206

 


Site contents and design Copyright 2006-16©