Treatment of a field case of avian intestinal spirochaetosis caused by Brachyspira pilosicoli with tiamulin

There has been much confusion over the significance of spirochaetes found in the caeca of laying hens and the impact they may have on egg production. In recent years, the situation has been made clearer and the presence of such species as Brachyspira pilosicoli have been shown to cause a mild, chronic disease in both layers and breeders and to reduce egg production by reportedly 5%.

1 June 2006

21 minute read

In the United Kingdom, a multi-age caged laying site with three separate flocks of approximately 12 000 birds each was chronically infected with B. pilosicoli but displayed few clinical signs except for a noticeable reduction in egg production and an increased mortality. The flocks were treated for 3 days in the drinking water with tiamulin at 12.5 mg/kg bodyweight, and a steady improvement in performance was recorded. The production results were compared with a flock that had been untreated with tiamulin previously, as a control, and one that had been treated at 25 and 45 weeks of age. A 9.8% improvement in egg production/hen housed up to 72 weeks of age and 9.7% in total egg weight was recorded, as well as an 8.6% reduction in actual hen mortality, in the tiamulin-treated flock in comparison with the untreated control. After taking into account the difference in breeds used, there was only a 6% reduction in egg production but an 8.84% increase in mortality in the untreated flock compared with the individual breed’s standard production data. The cost of the disease was estimated at £14 million in the United Kingdom, based on a national laying flock of 30 million or 1.5% of production. Faecal examination for potentially pathogenic spirochaetes should be part of the differential diagnosis of under-performing laying flocks.

Introduction:

Avian intestinal spirochaetosis is defined in general terms (Swayne, 2003), as a spirochaete colonization, predominately of the caecum and/or rectum of poultry. However, this is to be distinguished from the clinical disease, which is described in pullets, layers and breeders as causing soft to watery, brown to grey diarrhoea and pasty vents, and is often associated with a delayed onset of lay in pullets, reduced egg production and body weight in layers and poor chick quality from infected breeders (Smit et al ., 1998). The severity of the condition depends on the pathogenicity of the spirochaetes and their relative destruction of the surface epithelium of the caecum and degree of typhlitis caused. It is a chronic infection, with colonization and excretion of the organisms for several months (Dwars et al. , 1990; Hampson & McLaren, 1999).

Spirochaetes were commonly found in the caecum of layers and breeders (Dwars et al. , 1989; Stephens & Hampson, 1999) but were not always associated with clinical disease. Initially, this led to confusion over the significance of spirochaete infections in chickens*until the 1990s, when the different spirochaetes found in poultry were better classified genetically and phenotypically (McLaren et al. , 1997; Phillips et al. , 2005). Stephens and Hampson (2002a) showed that Brachyspira pilosicoli was potentially pathogenic in breeder layers, causing a delayed onset of lay and a significant reduction in egg production, in comparison with Brachyspira innocens, which was shown to be non-pathogenic. Hampson & McLaren (1999) found that an isolate of Brachyspira intermedia also caused a significant reduction in egg production and also mean egg weight. Brachyspira alvinipulli has also been shown to be mildly pathogenic in the United States (Swayne et al ., 1992; Swayne, 2003).

The association of spirochaetal infections in laying flocks with enteric disease in Holland was reported by Dwars et al (1989). In a study of 179 flocks, spirochaetes could be demonstrated in 27.6% of flocks with reported enteric disorders but only in 4.4% of flocks with no enteric signs. In Australia (Stephens & Hampson, 1999), the prevalence of spirochaetes in 50 flocks tested appeared to be much higher; with 42.9% of breeder flocks and 68.2% of layer flocks infected using a different methodology. Isolates from a subset of 16 flocks were investigated to a species level, and B. pilosicoli and B. intermedia were found in eight flocks (50%) B. pilosicoli in six flocks (37.5%) and B. intermedia in four flocks (25%) either alone or in mixed infections. Bano et al. (2005) reported on a survey in northeastern Italy involving 29 layer farms. Spirochaetes were found in 72.4% of the layer farms and 71.1% of the sheds. The pathogenic spirochaetes B. pilosicoli and B. intermedia were found in 31% of the sheds in an approximate 1:2 ratio. The susceptibility of avian spirochaetes to a variety of antimicrobials was reported by Hampson & Stephens (2002) using an agar dilution method. Tiamulin was shown to be the most active, with 15/19 (79%) isolates having a minimum inhibitory concentration of less than 0.1 mg/ml and the remainder were between 0.1 and 1.0 mg/ml. Other anti-spirochaetal antimicrobials were also shown to be active, such as metronidazole, lincomycin and, to a lesser extent, tylosin. Stephens & Hampson (2002b) carried out an artificial challenge study with B. pilosicoli in broiler breeders and administered tiamulin at 25 mg/kg bodyweight/day by crop tube for five consecutive days. Birds that had been infected and treated stopped shedding B. pilosicoli, whereas the untreated infected birds continued to shed until termination of the study, 4 weeks later. It is the purpose of this report to describe the use of tiamulin in the field in flocks that had been chronically infected with B. pilosicoli and their response to treatment.

Clinical Case

A commercial laying farm, with three caged flocks of approximately 12 000 birds each was reported to be showing reduced egg production and poor performance over an 18-month period. Each flock was subjected to serological examination at 20 weeks of age to monitor the efficiency of the vaccination programme during rearing to act as a baseline for any subsequent investigations. Generally these point-of-lay results indicated good to very good levels of protection against the common viral infections. The subsequent production problems were investigated serologically using both haemagglutination inhibition and enzyme-linked immunosorbent assay tests, looking for challenges from infectious bronchitis virus (M41) and variants (793B, D274 and D1466), avian rhinotracheitis virus, Mycoplasma gallisepticum and Mycoplasma synoviae. There was evidence of rising titres from intermittent challenges by infectious bronchitis variants, mainly 793B and D274, in common with the situation on other units belonging to the company, but these did not appear to correlate with the production problems on this site as there were no clinical signs noted. All flocks had been vaccinated at least twice against the infectious bronchitis variants found. M. synoviae, which is commonly reported (79%) in the UK laying flocks (Hagen et al. , 2004), was also present in these flocks. There was no response to treatment with chlortetracycline administered via the feed at 600 parts/106 at regular 7 to 8 week intervals to control any mycoplasmal infections and related conditions. The level of soft faeces was higher than normal but not considered widespread. Faecal samples were collected in each shed from 12 cages, each containing five hens, by placing cling-film-covered cards under the cages. Aliquots of faeces, including any abnormal droppings were then transferred into six 5-ml containers, filled to capacity to exclude air. These samples were sent for examination at the Veterinary Laboratories Agency (VLA) in Sutton Bonington, specifically for Brachyspira spp. examination, which is not routinely carried out on poultry faeces in the United Kingdom. Spirochaetes were observed weakly fluorescing, following fluorescent antibody staining (fluorescent antibody test(FAT)) using a B. hyodysenteriae-specific antibody and positive samples were submitted to culture. Spirochaetes were subsequently cultured and shown to cause a weak haemolysis on trypticase soy agar plates with defibrinated sheep blood with antimicrobials, spectinomycin, colistin and vancomycin to prevent overgrowth by other bacteria. Cultures were sent to VLA Winchester for definitive identification by biochemical tests. The presence of B. pilosicoli was confirmed by biochemical differentiation and found to be indole-negative but hippurate-positive. Serum agglutination tests with B. hyodysenteriae- specific serum were negative. Initially, only one shed was shown to be positive, but on repeated sampling all three sheds yielded B. pilosicoli . Polymerase chain reaction tests for Brachyspira spp. were not available in the VLA in England and minimum inhibitory concentration (MIC) testing of isolates, especially from poultry, was not routinely carried out.

Several postmortem examinations were performed over the problem period, especially when mortality rates were much higher than usual. The vast majority of deaths were associated with peritonitis caused by Escherichia coli .

Following consultation, it was decided to treat the oldest flock with tiamulin at 12.5 mg/kg bodyweight for 3 days via the drinking water. The daily dose in pigs for B. hyodysenteriae, a related organism, which causes swine dysentery, was 8.8 mg/kg bodyweight for 3 to 5 days. It was reported that the birds showed consistent and steady improvements in faecal consistency, appetite, body weight, egg production and mortality. It was decided to medicate the two remaining younger flocks; one was 45 weeks of age and the other was 25 weeks of age. Additional tiamulin medication was applied at 60 weeks and 45 weeks of age to each flock, respectively, as a precaution against possible recurrence, following rigorous insect-control measures, as it was thought likely that flies might maintain the spread of the infection. Unfortunately, no further faeces samples were taken following medication to test for the elimination of the B. pilosicoli infection as it was a clinical problem in the field, rather than a clinical trial, and costs to the farmer were to be kept to a minimum.

Results

The production data from an earlier flock, which had not been treated with tiamulin, were compared with the two younger treated flocks (see Table 1).

A similar number of birds were housed in each house but the untreated control flock (Flock A) was of a different breed, Lohmann Brown, rather than the two treated flocks, which were Bovans Goldline (Flocks B and C).

In egg production terms the treated flocks showed a graded response to tiamulin treatment with the later treated flock (Flock B, 45 and 60 weeks), showing a 5.2% and 5.6% improvement over the untreated flock (Flock A) in egg numbers and total egg weight when adjusted to a 72 week age for comparison purposes, and the flock treated at 25 and 45 weeks (Flock C) giving a 9.8 and 9.7% improvement, respectively. The average egg weights were almost identical and only minor differences in egg quality were recorded. The laying curve (see Figure 1)

_{Table 1. Comparative production data for an infected untreated flock (Flock A), an infected flock treated with tiamulin at 45 and 60 weeks of age (Flock B) and a flock treated at 25 and 45 weeks of age (Flock C)}

Parameter	Flock A	Flock B	Flock C	Comparison of Flock C with Flock A (%)
Number of birds housed	12 402	12 032	12 156
Breed	Lohmann	Goldline	Goldline
Eggs/hen housed	290.53	304.88	320.33	10.26
Eggs/hen housed at 72 weeks^a	291.75	306.92	320.33	9.8
Egg weight (kg) at 72 weeks	18.59	19.63	20.40	9.7
Average egg weight (g)	63.71	63.96	63.68	-0.05
Feed consumption/day (g)	111.285	119.954	120.432	8.2
Total feed 20 to 72 weeks (kg)	40.50	43.66	43.83	8.2
Feed/egg conversion ratio	2.179	2.220	2.148	-1.42
Average bodyweight (kg) at sale	1.935	1.79	1.90	-1.3
Mortality (%)	13.84	7.11	5.25	-8.59
Peak laying (%)	91.5	90	94	2.5
Egg quality (%)
Very large	4.93	6.00	5.32	0.39
Large	40.27	40.36	42.05	1.78
Medium	38.08	34.12	34.33	-3.75
Small	4.94	3.56	4.35	-0.59
Seconds	11.78	15.96	13.94	2.16

_{^aAdjusted to 72 weeks of production for flock comparison data.}

shows that Flock C had a higher and longer peak laying percentage than the control A and B flocks but interestingly Flock B’s performance tended to follow Flock C’s performance following tiamulin treatment. The average feed intake of Flock A (control) was lower, which affected egg production; however, somewhat surprisingly, the average hen bodyweight in Flock A (control) was higher than both the Goldline flocks, and Flock B would be considered very low for the breed at that age. Mortality in Flock A was substantially higher than Flocks B and C by 6.73 and 8.59%, respectively. This would have had a marked effect on the flock’s overall performance when judged on a hen/ housed basis.

To take into account the breed variation, the laying flocks were compared with their own breed standards. Flock A (untreated Lohmann flock) was compared with the Lohmann Brown standard performance data and laying curve for caged birds (see Table 2 and Figure 2). Flocks B and C, which received tiamulin treatment, were compared with the Bovans Goldline caged-bird standard production figures (see Table 3 and Figure 3). Figure 2 demonstrated that Flock A started laying 2 weeks earlier than the standard and did not quite reach the normal peak of 94.5%, but only 91.5% (-3%). However, the major difference was the subsequent performance up to 72 weeks where there was a major divergence in production, and Flock Awas down to 60% and the standard was 77.6% (-17.6%). Overall, Flock A produced 18.7 (-6%) fewer eggs than the standard and almost 1 kg less in egg weight (-5.39%). Feed consumption was reduced by 2.03% in Flock A and the overall feed/egg conversion ratio (FCR) was worse by 3.56%. The most striking feature was the increased mortality rate of 13.84% in comparison with the flock standard of 5% (+8.84%). Flock C (treated at 25 and 45 weeks of age) by contrast, mirrored fairly closely the Goldline standard laying curve. Peak production was 1% lower at 94%, but otherwise it followed the laying curve right up to

Figure 1. Comparative laying curves of an infected untreated flock (Flock A), an infected flock treated with tiamulin at 45 and 60 weeks of age (Flock B), and a flock treated at 25 and 45 weeks of age (Flock C). T, tiamulin treatment of Flocks B and C.

_{Table 2. Comparative production data for a B. pilosicoli-infected untreated flock (Flock A) and the Lohmann Brown caged hen standard to 72 weeks of age}

Parameter	Flock A	Lohmann Brown standard data to 72 weeks	Comparison of Flock A with flock standard (%)
Breed	Lohmann	Lohmann
Eggs/hen housed at 72 weeks^a	291.75	310.4	-6.0
Egg weight (kg) at 72 weeks	18.59	19.65	-5.39
Average egg weight (g)	63.71	63.30	0.65
Feed consumption/day (g)	111.285	113.600	-2.03
Total feed 20 to 72 weeks (kg)	40.50	41.35	-2.06
Feed/egg conversion ratio	2.179	2.104	3.56
Average bodyweight (kg) at sale	1.935	1.900	1.84
Mortality (%)	13.84	5	8.84
Peak laying (%)	91.5	94.5	-3.0

_{^aAdjusted to 72 weeks of production for flock comparison data.}

65 weeks when there was a small drop in production by 72 weeks of 4%. Overall, Flock C laid 7.3 more eggs than the standard (+2.34%) and because the average egg weight was also higher, produced 0.8 kg (+5.15%) higher egg weight. Feed consumption was noticeably higher by almost 3 kg and the FCR was therefore worse by 1.8%. The mortality rate was similar to the standard, only 0.25% higher.

Flock B, which was treated midway (week 45) and at week 60, showed a lower productivity than standard throughout the laying period. Peak laying percentage was down by 5% from standard and this continued for the rest of the laying period. Egg production was 1.94% lower than standard (-6.1 eggs) but average and total egg weight was higher by 3.19 and 1.19%, respectively. Feed consumption was higher than standard and the FCR was worse by 5.21%. Mortality was 2.22% higher than standard.

The comparative differences of the effects of the disease and treatments with the breed effect removed are summarized in Table 4. There was a graded response to treatment, with the untreated birds (Flock A) showing a 6% reduction in the number of eggs produced per hen housed to week 72, the late-treated birds (Flock B) showed a 1.94% drop and the early treated birds (Flock C) a 2.34% increase. The drop in egg production of 6% in the untreated group is closer to the 5% drop described by Swayne (2003). There was also a graded response to total egg weight. With regard to total feed intake there was less of a graded effect but more of a breed effect, with Goldline flocks consuming substantially more. There was no graded effect on the FCR but both Flocks A and B had a higher FCRvariation then Flock C. The average bodyweight was very variable and breed related. Both the Goldline flocks were lower than the standard for the breed. There was a marked difference between the breeds at 72 weeks of lay, with Lohmann flocks meant to be 1.9 kg (declining from a high of 1.95 kg at 54 weeks) and Goldline flocks 1.99 kg bodyweight (increasing from 1.97 kg at 54 weeks). There was a graded response to mortality, with the untreated flock showing a substantially higher rate than the other two treated flocks. The peak laying percentage was lower in the untreated flock (Flock A) as well as Flock B, which was not treated until much later at week 45, than the early treated Flock C (25 weeks of age), which suggests the infection could well be having an impact on this parameter. Unfortunately, no statistical analysis of these data was considered possible due to the lack of replication of the flocks and treatments.

_{Figure 2. Comparative laying curves of the infected untreated flock (Flock A) and the Lohmann Brown standard.}

Discussion

B. pilosicoli is meant to be associated with a mild intestinal infection in laying birds with a drop of 5% in egg production and no impact on mortality (Swayne, 2003). However, if the condition is undiagnosed and untreated and lasts for the full length of the laying period, the effect seen was a 6% reduction in egg production and an increase of mortality by 8.84%. The mortality did not appear to be directly related to the B. pilosicoli infection, because it appeared to be associated with peritonitis caused by E. coli , but might be indirectly associated with the chronic debilitating stress of a chronic intestinal infection. It was not thought to be associated with the M. synoviae infection, although E. coli infections are commonly secondary to mycoplasmal infections, because this high level of mortality was not seen in other flocks, which also had M. synoviae infections and were controlled in the main by chlortetracycline medication. These factors demonstrate the chronic insidiousness of B. pilosicoli infections and highlight that, where there is a low grade level of diarrhoea, poor production and an increased mortality, spirochaetosis should be considered as a differential diagnosis.

Because chlortetracycline at 600 parts/106 in feed had no effect on the condition, in spite of repeated use, tiamulin was considered the next suitable alternative in layers, as it also has a zero withdrawal period for eggs in the United Kingdom. Tetracycline appears to have a moderate activity in vitro against B. pilosicoli , with an MIC90 below 5 mg/ml (Hampson & Stephens, 2002); but as gut pharmacokinetic and pharmacodynamic data are not available for chickens, a suitable microbiological breakpoint could not be determined and it is suspected that the organism was resistant to chlortetracycline. The dose of tiamulin used (12.5 mg/kg bodyweight) was lower than the one indicated for mycoplasma treatment in the United Kingdom (25 mg/kg bodyweight) and by Stephens & Hampson (2002b), but higher than the level used in pigs for B. hyodysenteriae infections (8.8 mg/kg bodyweight). There are also no data available for gut pharmacokinetics of tiamulin in chickens to establish a breakpoint, but a positive clinical response was achieved. Unfortunately, no faecal samples were taken directly after treatment to determine whether the organism had been eliminated from the gut, as it was a field case. However, previous work by Stephens & Hampson (2002b) showed that indeed the organism was promptly removed from the birds following treatment and did not return within the 4-week observation period. At necropsy the organisms were also not recovered, unlike in the untreated controls. The birds were kept in cages in this field study, which tend to reduce faecal contamination and reduce the opportunities for re-infection, unlike free-range or barn-reared flocks, but a further application of tiamulin was given some months after the initial treatment as a precaution against re-infection.

_{Table 3. Comparative production data for tiamulin-treated flocks (Flocks B and C) and the Bovans Goldline caged hen standard to 72 weeks of age}

Parameter	Flock B, treated weeks 45 and 60	Flock C, treated weeks 25 and 45	Bovans Goldline standard data to 72 weeks	Comparison of Flock B with Goldline standard (%)	Comparison of Flock C with Goldline standard (%)
Breed	Goldline	Goldline	Goldline
Eggs/hen housed at 72 weeks^a	306.92	320.33	313	-1.94	2.34
Egg weight (kg) at 72 weeks	19.63	20.40	19.40	1.19	5.15
Average egg weight (g)	63.96	63.68	61.98	3.19	2.74
Feed consumption/day (g)	119.954	120.432	112.362	6.76	7.18
Total feed 20 to 72 weeks (kg)	43.66	43.83	40.90	6.75	7.16
Feed/egg conversion ratio	2.22	2.148	2.110	5.21	1.8
Average bodyweight (kg) at sale	1.79	1.90	1.99	-10.05	-4.74
Mortality (%)	7.11	5.25	5.0	2.11	0.25
Peak laying (%)	90.0	94.0	95.0	-5.0	-1.0

_{^aAdjusted to 72 weeks of production for flock comparison data.}

_{Figure 3. Comparative laying curves of the tiamulin-treated flocks (Flocks B and C) and the Bovans Goldline standard. T, tiamulin treatment of Flocks B and C.}

The cause of the spread of the disease from flock to flock on this multi-age site was not established but was thought to be most probably caused by mechanical transmission, probably by flies but mice or other carriers could be involved. It was not affected by the season and continued all year round. Bano et al. (2005) also reported that it was more common (86%) in birds older than 40 weeks of age than birds between 20 and 40 weeks of age (47%), and this was also found by Stephens & Hampson (1999). Birds kept in sheds with a deep pit system for droppings had a prevalence of 89%, in comparison with 58% for a conveyor belt system and frequent removal.

It is interesting that two recent surveys in Australia and Italy (Stephens & Hampson, 1999; Bano et al. , 2005) have shown a much higher prevalence of the potentially pathogenic spirochaetes B. pilosicoli and B. intermedia than previously considered by the Dutch study (Dwars et al. , 1989); this could be explained by differences in the techniques used. Dwars et al. (1989) used a direct FAT on samples sent into the laboratory, whereas the others used a faecal sampling and culture technique followed by a polymerase chain reaction test for isolate identification purposes. Both the Australian and Italian surveys showed that the prevalence of spirochaetes was about 70% (68 to 72%) and approximately one-half of those farms had pathogenic isolates of B. pilosicoli and B. intermedia (34 to 31%, respectively). Swayne (2003) reported that the potential economic impact of the disease has not been estimated. However, in the United Kingdom, with a national flock of approximately 30 million hens and potentially 30% of flocks infected that have a depression of egg production of 5% and an egg value of 3 p/egg, then a figure of approximately £14 million can be estimated as the potential loss to the UK commercial laying industry, or 1.5% of production.

_{Table 4. Comparison of individual flock variations (%) from their breed standard}

Parameter	Flock A, infected and untreated	Flock B, treated weeks 45 and 60	Flock C, treated weeks 25 and 45
Breed	Lohmann	Goldline	Goldline
Eggs/hen housed at 72 weeks^a	-6.0	-1.94	2.34
Egg weight (kg) at 72 weeks	-5.39	1.19	5.15
Total feed 20 to 72 weeks (kg)	-2.06	6.75	7.16
Feed/egg conversion ratio	3.56	5.21	1.8
Average bodyweight (kg) at sale	1.84	-10.05	-4.74
Mortality (%)	8.84	2.11	0.25
Peak laying (%)	-3.0	-5.0	-1.0

_{^aAdjusted to 72 weeks of production for flock comparison data.}

References:

Bano, L., Merialdi, G., Bonilauri, P., Dall’Anese, G., Capello, K., Comin, D., Cattoli, V, Sanguinetti, V. & Agnoletti, F. (2005). Prevalence of intestinal spirochaetes in layer flocks in Treviso province, northeastern Italy. In Proceedings of the Third International Conference on Colonic Spirochaetal Infections in Animals and Humans (pp. 56 -57). Parma, Italy.

Dwars, R.M., Smit, H.F., Davelaar, F.G. & Van ‘t Veer, W. (1989). Incidence of spirochaetal infections in cases of intestinal disorder in chickens. Avian Pathology, 18 , 591 -595.

Dwars, R.M., Smit, H.F. & Davelaar, F.G. (1990). Observations on avian intestinal spirochaetosis. Veterinary Quarterly, 12 (1), 51 -55.

Hagan, J.C., Ashton, N.J., Bradbury, J.M. & Morgan, K.L. (2004). Evaluation of an egg yolk enzyme-linked immunosorbent assay antibody test and its use to assess the prevalence of Mycoplasma synoviae in UK laying hens. Avian Pathology, 33 (1), 93 -97.

Hampson, D.J. & McLaren, A.J. (1999). Experimental infection of laying hens with Serpulina intermedia causes reduced egg production and increased faecal water content. Avian Pathology, 28 , 113 -117.

Hampson, D.J. & Stephens, C.P. (2002). Control of intestinal spirochaete infections in chickens. Report for the Rural Industries Research and Development Corporation, RIRDC No 02/087 .

McLaren, A.J., Trott, D.J., Swayne, D.E., Oxberry, S.L. & Hampson, D.J. (1997). Genetic and phenotypic characterisation of intestinal spirochaetes colonizing chickens and allocation of known pathogenic isolates to three distinct genetic groups. Journal of Clinical Microbiology, 35 (2), 412 -417.

Phillips, N.D., La, T. & Hampson, D.J. (2005). A cross-sectional study to investigate the occurrence and distribution of intestinal spirochaetes (Brachyspira spp ) in three flocks of laying hens. Veterinary Microbiology, 105 , 189 -198.

Smit, H.F., Dwars, R.M., Davelaar, F.G. & Wijtten, G.A.W. (1998). Observations on the influence of intestinal spirochaetosis in broiler breeders on the performance of their progeny and on egg production. Avian Pathology, 27 , 133 -141.

Stephens, C.P. & Hampson, D.J. (1999). Prevalence and disease association of intestinal spirochaetes in chickens in eastern Australia. Avian Pathology, 28 , 447 -454.

Stephens, C.P. & Hampson, D.J. (2002a). Experimental infection of broiler breeder hens with the intestinal spirochaete Brachyspira (Serpulina ) pilosicoli causes reduced egg production. Avian Pathology, 31 , 169 -175.

Stephens, C.P. & Hampson, D.J. (2002b). Evaluation of tiamulin and lincomycin for the treatment of broiler breeders experimentally infected with the intestinal spirochaete Brachyspira pilosicoli. Avian Pathology, 31 , 299 -304.

Swayne, D.E. (2003). Avian intestinal spirochaetosis. In Y.M. Saif, H.J. Barnes, J.R. Glisson, A.M. Fadly, L.R. McDougald & D.E Swayne (Eds.), Diseases of Poultry, 11th edn (pp. 826 -836). Ames: Iowa State Press.

Swayne, D.E., Bermudez, A.J., Sagartz, J.E., Eaton, K.A., Monfort, J.D., Stoutenberg, J.F. & Hayes, J.R. (1992). Association of cecal spirochetes with pasty vents and dirty eggshells in layers. Avian Diseases, 36 , 776 -781.