CASE 3

An unusual case of Cannibalism in Broiler Parents

SUMMARY

On farm investigation of cannibalism in 21 week old broiler parents revealed a wide variation in body weight, variable feed intake and vent pecking by underweight birds. Post mortem evidence of histomoniasis was found in birds from the worst affected house only. The true contribution of the histomoniasis infection to the flock weight profile was not established. Extensive culling and management modifications improved flock performance.

THE COMPLAINT

The initial complaint was an increase in mortality due to cannibalism, in the form of vent pecking, affecting two houses on a four house broiler parent laying site. A site visit was arranged.

HISTORY

Rearing Site

24640 pullets and 2800 cockerels were placed on a four house site on 20/2/99.

House 4 was a large, new house, rearing approximately 13000 birds. House 1-3 were older houses, rearing approximately 5000 birds each.

Weight profile

1.      Pullets

Bodyweight fluctuated from being 19.6% over target weight at 14 days old to 5% below target at 8 weeks old.

CV (Coefficient of Variation) varied from 10% between 3 & 8 weeks of age to 7.8% at 17 weeks.

2.      Cockerels

Bodyweight fluctuated between 13.1% below target at 4 weeks old to 1.5% over target at 14 weeks old.

CV varied from 12% at 6 & 7 weeks to 8% at 17 weeks.

Weekly weighing of birds was by hand in houses 1 ,2 & 3. In house 4 an automatic platform weigher (Weltech Engineering) was used. Grading occurred at 2 & 5 weeks of age.

Mortality

Transfer details

Transfer occurred on 20/6/99 at 17 weeks 1 day old.

Culling of unsatisfactory birds occurred during selection at transfer.

The birds were reportedly nervous and flighty upon arrival at the laying site and culling of small birds increased in houses 3 & 4 from the first week after arrival. Cannibalism was then seen from 19 weeks old.

INVESTIGATIONS

FARM VISIT

Visit to Farm A 16.7.99

The Site

A four house site. All houses had a single central row of double sided autonests, with a raised slatted area. The remaining floor was deep litter. Pullets were fed from track feeders with flat wooden board cockerel excluders. Cockerels were fed from raised pan feeders. Houses 1 - 3 had bell drinkers, whereas house 4 had a nipple line with an electrified anti-perching wire. The drinkers were situated on the raised slatted area. Bird numbers were similar in all houses.

Observations

The birds were 21 weeks old at the time of the visit. Cannibalism was only reported in houses 3 and 4. The visit was timed to coincide with feeding at 6.30am. For the visit, the feeders were put on sequentially from houses 1-4, over a 1½ hour period, to allow observation of the mechanics of feeding, bird behaviour and to identify sick or poor birds.

In all houses the birds were observed just prior to, during and after feeding. Four slave hoppers were used along the feed track to speed up feed distribution. It took 2½ minutes for feed to be distributed around the whole length of the track, which was satisfactory.

In all houses, some cockerels were able to feed from the feed track and pullets were seen eating from the cockerel feeders.

In houses 3 and 4, a large number of small, underweight birds were seen distributed throughout the house. They were easily identified, standing facing into the corners or sitting along the wire below the slats, making no attempt to feed. After feeding, unthrifty birds were not seen on the slats but were seen wandering around the feeder track, pecking at the litter below the feeders.

In Houses 1 and 2, the crop size of the birds after feeding was very consistent and within ½ hour a significant proportion (approximately 1/3) of birds were seen surrounding the drinkers. In houses 3 and 4, crop fill was very inconsistent and the birds took a long time (approximately 1-1½ hours) for a reasonable proportion (approximately ¼) of birds to jump onto the slats to drink.

Fifteen birds in house 4 and thirty birds in house 3 were culled due to low bodyweights. These birds had either had not eaten or had eaten very small amounts of feed, as measured by crop filling. Two and four vent pecked birds were also culled in houses 3 & 4 respectively. These were birds in good bodily condition, with blood on their tail feathers.

In total four birds were seen in houses 3 & 4 with blood staining of their head and neck feathers, suggesting that they had been pecking birds. These birds were thin. Other underweight birds were seen either pecking or wandering aimlessly about the littered area after feeding.

Of the birds culled from houses 3 and 4 during the visit, 38 of the 46 culls had hooked beaks. 10 out of 20 birds randomly examined in houses 1 & 2 combined, had hooked beaks. This suggested that the presence of hooked beaks was not primary reason for the vent pecking, but would exacerbate the severity of wounds.

Air quality appeared satisfactory in all houses, based on smell and dust levels. Temperature was satisfactory at 23-25^oC at an outside ambient temperature of 25^oC. The numbers of drinkers were satisfactory. The depth of the water in the bells and also the pressure in the nipple lines were satisfactory. No regular feeder breakdowns had been reported on the site since transfer and none occurred during the visit.

Bodyweights

Target bodyweight at 20 weeks 5 days is 2355 gms. Measured bodyweights, taken two days prior to the visit at this age, were as follows:

The assistant farm manager suggested that the bodyweights were satisfactory, compared to target.

No CVs (coefficient of variation (%) ) were calculated, even though an illustrated method of calculation was present on the weight recording charts. This omission was discussed on farm. It was revealed that no CV measurements had been calculated since the birds were transferred.

The CVs calculated from the 20 weeks 5 days weights were as follows:

Target CV was <8-10% (Cobb Breeder Management Guide)

Weighing was carried out on a Wednesday morning, five days after feed increases were given. This was satisfactory.

The method of weighing was discussed and several problem areas were highlighted. If underweight birds were picked up during test weighing, they were culled out and not included in the sample weighing and only 40 pullets and 40 cockerels were weighed weekly. This represented 0.7% of pullets and 8.5% of cockerels per house. The target should be to weigh 5% of pullets and 10% of cockerels per week (Cobb recommendations) to ensure accuracy of information.

During the visit a test weighing was performed to allow observation of this procedure. A sample of 40 pullets were weighed by the assistant farm manager. The current procedure consisted of weighing ten birds at four separate positions within the house, starting on the left hand side immediately inside the entrance door, then halfway down the house on the left and right hand side and lastly immediately inside the entrance door on the right. The far end of the house was not sampled. The lights were not dimmed and birds were not penned up. This resulted in some degree of bias, since birds that did not run away were most likely to be weighed. These would tend to be heavier birds having eaten well. Small thin birds would tend to run away, unless severely emaciated or dehydrated. If underweight, birds were culled and not weighed.

Average bodyweights and CV’s calculated from 40 pullets in House 3 weighed during the visit were as follows;

Three birds were picked up which were emaciated, whose weights were not included in this average weight and CV, as was normal practice for this farm.

If these three culled birds were included, the weight profile changed as follows;

This was a marked difference to that seen in weighing two days previously and highlighted the massive variation in bodyweights within the house.

The average weight of birds culled/dead from Houses 3 and 4 during the visit were as follows:

This included vent pecked birds, which were on target weight for age, highlighting the emaciated nature of the culls.

The high mortality reported over the two or three days prior to the visit was due to culling of underweight birds. Very few dead birds had been picked up, suggesting cannibalism was not a large proportion of the daily mortality.

Water consumption in Houses 3 and 4 was approximately 10% lower than that in houses 1 and 2. This difference may have been related to environmental factors such as  temperature and humidity differences, or related to reduced water intake by underweight birds.

Recommended Actions

The following actions were recommended during the visit;

1.   A reduction in light intensity from approximately 25-30 to 15-20 lux in houses 3 and 4 had already been implemented by the farm management.

2.   More extensive culling of small birds. Increased culling during feeding was suggested, as underweight birds would be most easily identified at this time.

3.   Approximately 12 bell drinkers were to be placed on the floor in each of the affected houses.

4.   Grading and penning of moderately underweight birds was advised, but the house design would not allow even access to drinkers and feeders if pen divisions were used. Pens were not built.

5.   Aspirin medication was prescribed for three days.

Initial Discussion

The cannibalism seen on this site was due to the wide variation in bodyweight in houses 3 and 4. Underweight birds were initiating the pecking which was then continued by all the birds in the house. After culling, the daily feed allowance was to be recalculated as underweight birds were being allocated feed that they were not eating, which would further exacerbate the wide CVs measured. The above recommended actions were to improve flock CV, by improving the accuracy of feeding, and removing the underweight birds. Continued low light intensity and removal of affected birds would reduce new cases of cannibalism developing.

Aspirin medication was prescribed to reduce the cannibalism. It tends to have a soporific effect on birds, particularly when combined with other calming measures, such as reduced light intensity.

Beak trimming was not suggested, as it would be too stressful to birds, just prior to the onset of lay.

POST MORTEM EXAMINATIONS

Post Mortem Ref: P0795

6 culled, underweight birds from house 3 were returned to the laboratory for post mortem examination. These were not examined on site due to time constraints.

Weight range 1.1-1.3kg.

The birds were grouped into two categories on the basis of post mortem findings. Two had caecal cores, with white/cream onion like layers of debris filling the lumen. Multiple areas of poorly discernible focal pallor was seen in the livers. Microscopic examination of scrapes of the caecal mucosa proved negative for coccidial stages. Swabs of caecal cores were cultured for salmonella with negative results (24 hour incubation in buffered peptone water at 37^oC, followed by incubation at 42^oC in Rappaport Vassiliadis liquid media for 24 & 48 hours, followed by incubation at 37^oC on Brilliant Green agar). A provisional diagnosis of histomoniasis was made. Four birds were underweight with no obvious gross visible lesions.

Post Mortem Ref: P0804

Ten culled birds from each of houses 3 & 4 were requested in order further clarify the presence and assess the extent of the histomoniasis infection.

House 3

Weight range 1.1-1.28kg.

Caecal and liver lesions, grossly resembling classical histomoniasis infection were seen in four  birds. Caecal cores were seen alone in a single bird. No obvious lesions were seen in the five remaining birds. In total six birds were dehydrated.

House 4

Weight range 0.82-2.54kg.

No obvious lesions were seen in any of the ten birds submitted. Four birds were dehydrated.

The significance of the histomoniasis in contributing to the weight loss/inadequate weight gain was not known. This protozoal infection can lead to weight loss. However it may be secondary to the poor condition of the birds, related to reduced immunity, imbalance of gastrointestinal flora or increased protozoal burden due to extensive litter eating.

Post Mortem Ref: P0816

Ten culled, on target bodyweight birds from each of houses 3 & 4 were requested in order to further evaluate the extent of infection within houses 3 & 4.

House 3; Weight range 2.28-2.8kg

House 4; Weight range 2.32-2.88kg

No gross lesions resembling histomoniasis infection were seen in any birds.

Heterakis Worm Count

Composite faeces and caecal contents from the submitted birds were analysed for worm eggs and adult worms using the Modified McMaster technique and a method of direct identification. Composite caecal contents were dissolved in approximately 150mls of distilled water, by repeated agitation over several days. The resultant suspension was poured onto a 150um sieve, and the sievings examined for adult worms.

Histopathology

Fixed caecum and liver were submitted for histopathological examination.

Three caecal sections had a severe necrotising typhilitis. Masses of purulent debris and bacteria were seen in the caecal lumen. A mononuclear cell infiltrate was seen diffusely infiltrating the mucosa and muscularis. Poorly staining round structures in the mucosa were suspicious of histomonads.

Coalescing areas of acute hepatocyte necrosis, with associated inflammatory cell infiltration were seen, together with large numbers of autolysing protozoa.

These findings were suggestive of histomoniasis infection.

DIAGNOSIS

Cannibalism subsequent to uneven weight gain post transfer. Histomoniasis may have contributed to or resulted from the reduced weight gain.

Medication

No medication was prescribed for this flock. Water soluble dimetridazole (Emtryl;Merial) has been successfully used to treat outbreaks of histomoniasis in broiler parents previously. However this product was no longer legally available for treatment of broiler parents, due to changes in licensing, which occurred on 30^th April 1999. If it had been available, it would have been recommended in house 3 on the basis of uneven weight gain, and post mortem findings.

DISCUSSION

Post transfer, weight gain in the birds in houses 3 & 4 became uneven. Immediately after transfer, these birds were reportedly not willing to jump on to the slats and drink after feeding. The fact that only two out of four sheds were affected on the site, suggested that this problem was a result of rearing management or environment rather than a laying site problem. The layout of the laying houses differed in drinker type only, with house 4 having nipple drinkers with an electrified anti-perching wire.

The two affected houses were reared in a single 13000 bird pen-less house on pan feeders and nipple drinkers. It was observed in rear that there were a lot of smaller birds in this house, which were culled out.

The reluctance of the birds to drink suggests that they did not quickly find the raised bell drinkers in house 3, or had an aversion to the nipples in house 4, possibly due to the electrified wire, which was not present on the rearing site drinkers. If there was a wider variation in bodyweight than suggested by the recorded CV, then small birds would be not be able to compete for feed and water.

The birds in houses 1 & 2 were reared in different accommodation to houses 3 & 4. Three houses with pens, bell drinkers and floor feeding were used. There was no evidence of uneven bodyweights, cannibalism or histomoniasis in these birds.

Calculation of CVs together with average bodyweights on a weekly basis must be recorded to give a picture of the progression of the flock. This was not performed on this site.

Hooked beaks increase the severity of pecking wounds and result from lack of wear, often related to feed and drinker type and bird health. Only cockerels were beak trimmed at 3 days old. In houses 1 and 2 hooked beaks were observed in a large number of birds, although not as hooked or as numerous as in houses 3 and 4. This may be related to the uneven bodyweights seen in houses 3 & 4, as a result of reduced feed and water intake. Because cannibalism was not seen in houses 1 & 2, this suggests that other factors precipitated the pecking, but the severity was increased due to the hooked nature of the beaks.

Histomoniasis can result in chronic weight loss or poor weight gain, resulting in mortality due to culling or chronic dehydration. Histomoniasis was only seen in house 3. Within house 3 it was limited to approximately 40% of the severely underweight birds submitted for post mortem examination. Caecal lesions can be seen from 8 days post infection and liver lesions can be seen from 10-12 days after infection (McDougald 1997). Post mortem confirmation of infection occurred at 21 weeks old, three weeks after transfer. It is difficult to assess the point of onset of infection. Histomonads use heterakis worms as intermediate hosts. Heterakis eggs are very persistent. Earth worms can be transport hosts. All rearing and laying site houses had concrete floors, reducing the risk of infection. Overt signs of histomoniasis can be apparent from 7 days post infection. Unevenness was seen from 19 weeks old. If this was related to histomoniasis infection, rather than environmental factors, then the infection most likely occurred on the rearing site. In contrast, only birds in house 3 were diagnosed with this infection, when houses 3 & 4 were reared in the same house. This would highlight the laying site as the source. However house 4 birds were placed before house 3. Small birds may crowd to the back of the rearing houses at transfer, leading to accumulation in house 3. Post mortem examination of underweight birds at or immediately after transfer may have clarified the timing of infection.

The laying potential of this flock was of concern, due to the wide CV’s observed. Overweight birds can contribute to higher mortality between point of lay and peak lay, due to loss of ovarian hierarchy, large egg size leading to cloacal prolapse, peritonitis, and heat stress in summer months. Underweight birds are unlikely to be in the appropriate body condition to respond to the fixed timing of feed and light increases, which are used to stimulate onset of ovarian maturity. Extremely underweight birds are likely to be out-competed for food and water, and will be culled. In summary, accurate weight assessment and feeding is necessary to control mortality, egg weight and egg production.   The reduction of light intensity to control pecking can also affect performance if extreme, by reducing light stimulation required for egg production.   The intensity was reduced to approximately 15-20 lux, which is above the minimum breed recommendation of 15 lux.

The poor adaptation of the birds in houses 3 & 4 to the new environment was the biggest contributing factor to poor flock performance post transfer. The transition from rearing accommodation to the laying house was problematic, due to the birds being unwilling to mount the slats to drink. The exact cause of this unwillingness was not identified. Extra efforts made at this stage to regularly move birds onto the slats were not made. This resulted in uneven weight gain and cannibalism. The movement of birds between these two types of accommodation in rearing and laying was to be reviewed. The introduction of perches and raised slatted areas into the rearing accommodation was suggested, to familiarise the birds to the laying house environment.

The contribution of the histomonas infection, seen in house 3 birds, to the poor performance was not established. However the variation in body weight and mortality was more severe in house 3, suggesting that it contributed in part to the observed problems.