Morphometric evaluation of “dysbacteriosis” in broilers

Abstract

In consequence of the withdrawal of products that assisted animal production, such as antimicrobial growth promoters, once-controlled enteric diseases have returned and new multifactorial diseases causing gut disorders of unknown origin have emerged in broilers. One of these widespread syndromes causing intestinal health problems in broilers is in the field referred to as “dysbacteriosis”. During the present study, the histopathology of the intestinal tract of broilers affected with dysbacteriosis was analysed. Commercial broilers were given a macroscopic dysbacteriosis score by experienced veterinarians during necropsy. Samples from the duodenum and caecum were taken from each broiler for histopathological analysis. An increase in the macroscopic dysbacteriosis score coincided with increased villus atrophy, a decrease in the thickness of the tunica muscularis and an increase in T-lymphocyte infiltration in the gut mucosa. Also more and larger goblet cells were observed in the animals with high macroscopic dysbacteriosis scores. Although the exact aetiology still remains to be identified, dysbacteriosis in broiler chickens thus coincides with an inflammatory reaction in the gut mucosa.

Introduction

Since the ban of antimicrobial growth promoters in Europe, the broiler industry has been facing a rise in intestinal health problems, collectively referred to as “dysbacteriosis” among practitioners (Eshuis et al., 1998; Fabri, 2000; Rebel et al., 2006; De Gussem, 2007). Dysbacteriosis has been defined as the presence of a qualitatively and/or quantitatively abnormal microbiota in the proximal parts of the small intestine, inducing a cascade of reactions in the gastro-intestinal tract including reduced nutrient digestibility and impaired intestinal barrier function, increasing the risk of bacterial translocation and inflammatory responses (Fabri, 2000; Panneman, 2000; van der Klis & Lensing, 2007).

The syndrome is generally seen between 20 and 30 days of age (Fabri, 2000; Pattison, 2002; Wilson et al., 2005). Clinically, the main Signs are: pale, glistening or orange droppings with undigested food particles, wet and greasy droppings and hence dirty feathers, sometimes foamy caecal droppings, reduced physical activity, increased water intake, a decrease in feed intake with a check in weight or reduced gain rates and an increased feed conversion (Fabri, 2000; Pattison, 2002; Wilson et al., 2005; De Gussem, 2007). At necropsy, the main observations are thin, fragile, often translucent intestinal walls, watery or foamy intestinal contents and frequent orange mucus and undigested particles in the intestines, ballooning of the gut and intestinal inflammation (Pattison, 2002; De Gussem, 2007).

It is believed that both non-infectious and infectious factors can play a role in dysbacteriosis (Mortimer, 2002; De Gussem, 2007). Suspected non-infectious factors are different types of non-specific stressors, such as feed interruptions or dietary changes, nutritional imbalance, dietary stressors such as soluble ‘non-starch polysaccharides’, management disorders, genetic background, enzymatic dysfunction, and mycotoxins (Langhout et al., 1999; De Gussem, 2007; Teirlynck et al., 2009). Infectious agents that potentially play a role in dysbacteriosis are coccidia, Clostridium perfringens and other unidentified bacteria producing toxic metabolites (Morrow, 2001; De Gussem, 2007).

Despite the widespread nature and importance of the syndrome, there is a lack of scientific literature and still much controversy about the exact aetiology—and up until today the basis of the underlying pathophysiology is unknown. Hence, diagnosis in broilers faces a lot of challenges due to the incomplete characterization of the syndrome (Wilson et al., 2005).

The purpose of the present study was to gain insight into the histopathological changes at the level of the intestinal mucosa in field cases of dysbacteriosis.

Materials and Methods

Animals and sampling

Male Ross 308 broiler chickens were used in the study: 12 birds 18 days old (two flocks, one broiler farm), 38 birds 21 days old (four flocks, two broiler farms), 36 birds 28 days old (two flocks, one broiler farm) and 18 birds 42 days old (one flock, one broiler farm). In total, broilers of nine flocks from five different broiler farms were sampled. The flocks were selected adt random from an integrated poultry enterprise, thus all the birds originated from the same breeder flocks and were given similar feed including anticoccidials. The birds were receiving nicarbazin/narasin in the starter phase (day 1 until day 18) and salinomycin in the grower phase until 30 days of age at registered doses. This enterprise was reported by the field veterinarian to have above-average intestinal health problems, including dysbacteriosis, coccidiosis (mainly Eimeria maxima) and wet litter, although the (ventilation) management was evaluated as better than average. The birds that were selected were not clinically ill, and were considered as average, healthy birds by an experienced poultry veterinarian.

The chickens were euthanized by means of cervical dislocation by an experienced veterinarian and immediately necropsied. Intestinal parameters were scored (see below). Samples of approximately 3 cm were taken from the second limb of the duodenum and the middle part of one caecum, rinsed in phosphate-buffered saline and fixed in 4% (v:v) buffered formalin.

Macroscopic scoring system

Each bird was given a score between 0 and 10 for intestinal dysbacteriosis parameters, 0 being a normal gastrointestinal tract and 10 being the most severe dysbacteriosis. In total 10 parameters were assessed and scored 0 when absent and 1 when present (Figure 1). The parameters are: (1) ballooning of the gut; (2) significant redness of the serosal and/or mucosal side of the gut and/or presence of abnormally dilated blood vessels on the serosal side of the gut, cranial to Meckel's diverticulum; (3) a macroscopic visible and/or tangible reduction of the gut wall thickness and/or translucent guts in combination with increased fragility of the gut cranial to Meckel's diverticulum; (4) 3 sec after dissecting the gut, the edges of the gut cranial to Meckel's diverticulum are flaccid; (5) an abnormal appearance of the contents in the lumen of the gut (excessive slime, water, gas, greasy aspect or mixture of these) cranial to Meckel's diverticulum; (6) significant redness of the serosal and/or mucosal side of the gut and/or presence of abnormally dilated blood vessels on the serosal side of the gut, caudal to Meckel's diverticulum; (7) a macroscopic visible and/or tangible reduction of the gut wall thickness and/or translucent guts in combination with increased fragility of the gut caudal from Meckel's diverticulum; (8) 3 sec after dissecting the gut, the edges of the gut caudal to Meckel's diverticulum are flaccid; (9) abnormal appearance of the contents in the lumen of the gut (excessive slime, water, gas, greasy aspect or mixture of these) caudal to Meckel's diverticulum; and (10) undigested food particles caudal to ileo-caecal junction (De Gussem, 2010). During this lesion scoring, no intestinal sections were given scores of 0, 9 or 10.

Figure 1.  Macroscopic dysbacteriosis score system parameters. 1A: Overall gut ballooning. 1B: Content of the intestinal tract: (B1) mucoid, orange intestinal content; and (B2) foamy intestinal content. 1C: Tonus of the intestinal tract: (C1) good tonus; and (C2) lack of tonus. 1D: Macroscopically visible thickness of the intestinal tract: (D1) macroscopically thin intestinal tract; and (D2) intestinal tract with normal thickness. 1E: Undigested particles in the colon (arrows). 1F: Inflammation of the gut: (F1) inflammation; and (F2) no inflammation.

Morphological examination

Formalin-fixed intestinal segments were dehydrated in xylene, and embedded in paraffin. Sections of 4 µm were cut using a microtome (Microm; Prosan, Merelbeke, Belgium). Deparaffinization was done in xylene (2×5 min). The sections were then rehydrated in isopropylene (5 min), 95% alcohol (5 min) and 50% alcohol (5 min) and were stained with haematoxylin and eosin. Histological lesions were studied using standard light microscopy. Villus length in the duodenum was measured by random measurement of nine villi per section using an Olympus BX61 Digital Camera DP50 (Olympus NV, Aartselaar, Belgium) and a computer-based image analysis system (Analysis^® J-2; P4 Technologies, Inc., Waldorf, Maryland, USA). Only intact villi were measured, meaning villi for which the tip as well as the base of the villus were in the plane of the section. The thickness of the tunica muscularis in duodenum and caeca was also measured using the Analysis^® J-2 software. For each section, eight measurements were performed on different locations. Measurements were done on cross-sections of ring-shaped intestinal segments that allow unbiased perpendicular measurements. Detection of goblet cells was done using periodic acid Schiff staining as described by Forder et al. (2007)

Immunohistochemical examination

Deparaffinization of paraffin-embedded tissue sections (4 µm) was performed as described above. The pressure cooker antigen retrieval method (Tender Cooker, Nordic Ware, Minneapolis, USA) was applied to the samples. Immunohistochemical labelling of leukocytes was performed as described by Mast et al. (1998). Briefly, endogenous peroxidase in the tissue sections was blocked with H₂O₂ (3%) in methanol for 30 min (21°C). After rinsing with phosphate-buffered saline sections were incubated for 1 h (21°C) with monoclonal antibodies directed against T lymphocytes (KUL05). After rinsing thoroughly, a goat anti-mouse IgG₁ conjugate, labelled with peroxidase (Dako, Heverlee, Belgium), was added for 30 min (21°C). After rinsing again, tissue sections were incubated with ABC horseradish peroxidase complex (Dako, Heverlee, Belgium) for 30 min (21°C). After another rinse, positive cells were stained brown after conversion of the substrate (3,3′-diaminobenzidine tetrahydrochloride; Sigma, St Louis, Missouri, USA) in the presence of H₂O₂. The number of T lymphocytes in the propria mucosae was scored with an automatic image analysis system (Optimas 6.5; Media Cybernetics, Silver Spring, USA), measuring the area percentage occupied by the labelled cells. For each section, eight randomly selected sites were analysed by the image analysis program.

Statistical analysis

The relation between the macroscopic lesion score and the outcome variables villus length, thickness of the tunica muscularis and T-lymphocyte infiltration was evaluated by means of a linear mixed model taking into account the chicken as a random factor since multiple observations were made on one animal. In addition, other co-variables such as age of the bird, the person scoring the bird and the intestinal segment were also taken into account. First, all independent variables were tested univariably, and subsequently all significant variables were included in a multivariable model. In this multivariable model, all two-way interactions were also tested. In the multivariable model, all main effects and interactions with P<0.05 were retained. For the outcome variables that were not normally distributed, a log transformation was performed. Model fit was evaluated by means of evaluation of the residuals. The number of chickens belonging to a certain macroscopic dysbacteriosis score group per age group is presented in Table 1.

Table 1.  Number of chickens analysed per age per macroscopic dysbacteriosis group.

	Macroscopic “dysbacteriosis” score
Age	1	2	3	4	5	6	7	8
18 days	–	1	4	4	3	–	–	–
21 days	–	3	4	13	4	8	5	1
28 days	1	3	–	3	13	6	6	4
42 days	1	1	1	2	4	6	1	2
Both the duodenum and caecum were analysed for each chicken. –, no birds of this age group were given this score.

Results

Intestinal morphology

Villus length

Duodenal villus length generally decreased with an increase in the macroscopic “dysbacteriosis” score within one age group (Table 2). Statistical analysis showed that, starting from macroscopic dysbacteriosis score 5, the villus length was significantly lower compared with macroscopic dysbacteriosis score 1 (P=0.01). Score 8 differed significantly from scores 5, 6 and 7 (P=0.03).

Table 2.  Length of villi (µm) in duodenal sections in relation to the macroscopic score.

	Macroscopic “dysbacteriosis” score
Age	1	2	3	4	5	6	7	8
18 days	–	1988±153	1668±219	1627±188	1550±201	–	–	–
21 days	–	2125±260	1929±239	2006±292	1870±219	1694±54	1491±193	1495±126
28 days	2656±245	1866±152	–	2173±269	1878±290	1812±415	1762±336	1479±212
42 days	2439±151	2471±81	2847±222	2364±185	2049±208	2078±289	2527±190	1862±180
Random measurements of nine villi per three gut sections were performed by a computer-based analysis system. The given values ± standard deviations are the means of nine villi measured in the chickens from the same age and in the same macroscopic “dysbacteriosis” score category. –, no birds of this age group were given this score.

Cystic crypts

In the duodenum, a mild to severe dilatation of the crypts of Lieberkühn was observed. Although not quantitatively analysed, the number of cystic crypts appeared to increase with age and with the macroscopic dysbacteriosis score. The age effect seemed to have the highest impact.

Thickness of the tunica muscularis

For each age interval, the thickness of the tunica muscularis generally decreased with an increasing macroscopic dysbacteriosis score (Table 3). Statistical analysis showed that the thickness of the tunica muscularis of the gut of the animals having macroscopic dysbacteriosis score 5 and higher was significantly lower as compared with score 1 (P=0.01).

Table 3.  Thickness of the tunica muscularis (µm) in duodenal and caecal sections in relation to the macroscopic score.

	Macroscopic “dysbacteriosis” score
Age	1	2	3	4	5	6	7	8
18 days	–	197±22	152±28	143±25	119±16	–	–	–
21 days	–	233±40	186±30	178±39	174±33	144±16	144±27	127±23
28 days	293±76	306±105	–	253±62	211±52	185±37	170±26	151±23
42 days	327±61	329±91	277±45	246±77	205±29	193±45	195±21	165±30
Eight random measurements of the thickness of the tunica muscularis per three gut sections were performed by a computer-based analysis system. The given values ± standard deviations are the means of eight random measurements of both the duodenum and caecum of the chickens in the same age and in the same macroscopic “dysbacteriosis” score category. –, no birds of this age group were given this score.

Goblet cells

Although not quantitatively analysed, periodic acid Schiff staining generally showed more and larger goblet cells both on the villi and in the crypts of the gut of animals with higher macroscopic dysbacteriosis scores.

T-lymphocyte infiltration

Generally T-lymphocyte infiltration in the duodenum and caecum increased within one age group, with increased macroscopic dysbacteriosis scores (Table 4). Statistical analysis showed that T-lymphocyte infiltration in the gut mucosa of animals having a macroscopic dysbacteriosis score of 6 and higher was significantly higher as compared with score 1 (P=0.02).

Table 4.  T-lymphocyte infiltration (area percentage) for duodenal and caecal sections in relation to the macroscopic score.

	Macroscopic “dysbacteriosis” score
Age	1	2	3	4	5	6	7	8
18 days	–	4.72±1.70	5.64±2.36	5.54±2.14	6.52±3.31	–	–	–
21 days	–	4.06±1.70	5.05±2.01	5.56±2.64	5.10±2.35	6.30±3.10	6.80±3.05	8.20±1.95
28 days	5.59±1.80	7.93±3.88	–	7.70±3.19	7.48±2.83	10.24±3.19	8.74±3.66	11.00±3.06
42 days	7.82±2.35	6.97±1.47	8.44±2.20	5.57±1.72	7.24±2.58	8.61±2.71	7.69±1.33	9.50±2.01
Eight random measurements of the T-lymphocyte infiltration in the mucosa per three gut sections were performed by a computer-based analysis system. The given values ± standard deviations are the means of the eight random measurements of both the duodenum and caecum of the chickens in the same age and in the same macroscopic “dysbacteriosis” score category. –, no birds of this age group were given this score.

Discussion

Definitive diagnosis of dysbacteriosis has been challenging due to the non-specific nature of the clinical signs and lesions and because the aetiology is still unknown (Wilson et al., 2005). The present study has identified several more or less characteristic changes, which may aid in confirming the diagnosis of “dysbacteriosis”, and which help in differentiating this entity from other intestinal disorders, such as malabsorption syndrome (MAS) and runting–stunting syndrome (RSS).

MAS has been described as a gastro-intestinal disease affecting broilers during the first 2 weeks post-hatch (Zekarias et al., 2002; van Hemert et al., 2004). The main clinical signs of MAS are weight gain depression with non-uniform growth, stunting, diarrhoea with undigested food particles resulting in wet litter, retarded and defective feathering, pigment loss, distended abdomens, depression and early mortality. At necropsy, lesions are found in the digestive organs, especially the small intestine, which is grossly pale and distended, with mucoid contents, and there is lower mineralization of the thigh bones (Kouwenhoven et al., 1978a, b; Bracewell & Randall, 1984; Reece et al., 1984; Szabo et al., 1989; Reece & Frazier, 1990; McNulty & McFerran, 1993; Sályi & Glávitis, 1999; Songserm et al., 2002a, b; Van Hemert et al., 2004; Rebel et al., 2006). Histologically, intestinal lesions such as cystic crypts of Lieberkühn and villus atrophy are observed (Rebel et al., 2004).

RSS typically affects birds within the first 2 weeks of life and can be defined as a syndrome in which a number of individuals in a flock appear considerably small due to delayed growth (Dufour-Zavala, 2005; Nili et al., 2007). Clinically, the main signs of RSS are immobility, increased feed conversion and poor performance, uneven growth and stunting, inappetance and excessive water consumption, diarrhoea, retarded and defective feathering, distended abdomens and bone defects (Vertommen et al., 1980a, b; Ruff, 1982; Calnek et al., 1997; Shapiro et al., 1997, 1998; Sályi and Glávits, 1999; Dufour-Zavala, 2005; Nili et al., 2007). At necropsy, lesions are found in the small intestine, which is pale and thin, almost translucent, containing undigested food particles.

A main difference between dysbacteriosis and MAS/RSS, of which the latter could be identical entities with different names, is that in field cases of dysbacteriosis the growth retardation due to poorer absorption of nutrients is probable but very often average growth will still be in line with breed standards, and also the homogeneity of the flock will typically not be dramatically affected as with RSS/MAS. Also the age of disease induction is lower (below 2 weeks of age) in MAS/RSS as compared with dysbacteriosis (3 to 4 weeks of age). Dysbacteriosis and MAS/RSS are clearly distinct from subclinical necrotic enteritis because no ulcerations in the gut are found.

In the dysbacteriosis field cases, the duodenal villus length, thickness of the tunica muscularis and T-lymphocyte infiltration in the mucosa of the duodenum and caecum were altered. No necrotic lesions were observed. This combination of morphological and inflammatory changes combined with the different clinical appearance compared with other syndromes (such as MAS/RSS) is in our opinion sufficiently unique and characteristic to classify this syndrome as a separate entity, although microscopic measurements of MAS/RSS samples are necessary for confirmation. The microscopic findings of dysbacteriosis may explain the clinical signs, being depressed growth, wet litter and clinical depression (Toskes et al., 1975; Kaldhusdal & Hofshagen, 1992; Hoerr, 2001). Based on the measurements, variations in intestinal distension between healthy gut and severe dysbacteriosis can explain the variations in thickness of the tunica muscularis. Thus ballooning of the gut and the flaccid aspect of the gut wall may be directly associated with reduced tonus/rigour of the tunica muscularis. Variations in intestinal distension, however, can only partly explain the reduction in villus length and increase in villus thickness as observed in the severe dysbacteriosis cases. Thus severe dysbacteriosis is associated with an absolute reduction in the available absorptive surface area. The change in the size and number of goblet cells could explain the mucous content of the intestinal tract of affected birds. Cystic crypt formation could be due to a vitamin deficiency as a secondary effect of the malabsorption caused by villus atrophy and increased T-lymphocyte infiltration (Klasing & Austic, 2003).

Dysbacteriosis is defined in human medicine as a condition characterized by a shift in the microbiota favouring abnormal populations of bacteria to predominate within the intestinal tract with minimal intestinal pathology (Dumitrasco et al., 1980; Schoorel et al., 1980; Sidorchuk & Bondarenko, 1984; Klemparskaya et al., 1987). Despite the definition of dysbacteriosis, the microbiota composition and possible microbiota shifts have not been studied and the name dysbacteriosis is not really substantiated. However, there are indications suggesting that the intestinal microbiota composition may play a role in this syndrome. These indications include the response to certain antibiotic treatments, the consistency of the droppings and the inflammatory nature of the intestinal lesions.

In conclusion, dysbacteriosis in broilers is characterized by villus atrophy, a decrease in the thickness of the tunica muscularis and an increase in T-lymphocyte infiltration in the gut mucosa. These histological observations can explain performance problems and macroscopic observations of the gut at necropsy, in broilers with dysbacteriosis.

Acknowledgements

The authors would like to thank Sarah Loomans, Christian Puttevils and Delphine Ameye for their skilful technical assistance. Dieter Vancraeynest and Maja Mariën (Alpharma) are acknowledged for their help with macroscopic lesion scoring. The present work was funded by grant S6169 of the Federal Public Service Health, Food Chain Safety and Environment.