Incidence, pathology and prevention of keel bone deformities in the ...

British Poultry Science Volume 45, Number 3 ( June 2004), pp. 320–330

Incidence, pathology and prevention of keel bone deformities in the laying hen R.H. FLEMING, H.A. MCCORMACK, L. MCTEIR

AND

C.C. WHITEHEAD

Bone Biology Group, Roslin Institute (Edinburgh), Scotland

Abstract 1. As a baseline study of the nature and incidence of keel deformities in laying hens, keel condition was examined in three different strains of hen from a total of 4 different caged environments (two commercial farms and two experimental farms). Incidence of keel deformity on farms in end of lay hens ranged from 2.6 to 16.7%. Only 0.8% of younger 15-week-old pullets had deformed keels. 2. Incidence of keel deformities was unchanged in 100 birds sampled from a free-range system compared to conventional caged siblings at the same farm. 3. Keel condition was also examined in 5 selected generations of a study involving the use of a bodyweight-restricted selection index for skeletal improvement. Divergent selection for skeletal characteristics decreased incidence of keel deformity and improved radiographic density (RD) in high bone index (BI) hens compared to low BI hens in all selected generations. Male high BI keels were also improved compared to low BI. Shear strength measured in normal keels in generation 6 (G6) of the genetic study was improved in high BI hens compared to low BI hens. For all hens in the genetic study, those with normal keels had stronger tibiotarsus and humerus breaking strengths than hens with deformed keels. 4. Histopathology of keels representative of different deformities showed the presence of fracture callus material and new bone in all cases. This establishes that deformities are a result of trauma and are not developmental in origin. 5. Ash contents of keels, tibiae and humeri showed no differences between hens with normal and deformed keels. There were no differences in indicators of collagen cross-linkage in other bones between hens with normal keels and those with deformed keels. 6. It is concluded that lack of bone mass is the underlying cause of keel fracture and deformity in laying hens, rather than qualitative changes in bone, and that genetic selection can improve keel quality and prevent deformity.

INTRODUCTION ‘Caged layer fatigue’, or osteoporosis in laying hens, leads to loss of structural bone and increased incidence of fracture at various skeletal sites by the end of the laying period (Whitehead and Fleming, 2000). Of all bones examined in end of lay hens, Gregory et al. (1994) found that 10.2% had old and healed breakages (principally the furculum, ulna and humerus) and a further 16.5% breakages occurred during depopulation and transport (most frequently the pubis and the keel). At this stage, Gregory and Wilkins (1989) observed that 31.4% of birds had broken bones, rising to 45.3% following removal from transport crates and hanging on to shackles prior to stunning. All fractures occurring up to this point should emphasise concerns for bird welfare. On the processing line, further bone

breakages occurred, leading to contamination of meat with bone splinters and highlighting the fragility of bones in laying hens. Prior to carcase processing, Gregory and Wilkins (1989) also observed that the most commonly broken bones were: the ischium (17.2% of birds), keel (12.8%) and humerus (10.8%). Low breast muscle mass in modern hybrid laying hens leaves the keel particularly vulnerable to fracture. Gregory and Devine (1999) scored protuberance of the keel and breast muscle condition to determine the extent of emaciation in end-of-lay hens. In the flock with the highest prevalence of low-scoring birds, they classed 9% of the birds as severely emaciated. ‘Crooked keel’ has been a long-standing problem in laying hens and has been attributed as variously, a hereditary disease (Warren, 1937), rachitic in origin (Winter and Funk, 1941), and due to faulty metabolism

Correspondence to: R.H. Fleming, Roslin Institute, Roslin, Midlothian EH25 9PS, Scotland. Fax: þ44-131-440-0434. E-mail: [email protected] Accepted for publication 21st November 2003.

ISSN 0007–1668(print)/ISSN 1466–1799(online) ß 2004 British Poultry Science Ltd DOI: 10.1080/00071660410001730815

KEEL DEFORMITIES IN HENS

and a ‘slowness’ to calcify (Buckner et al., 1946). Whether or not there is a hereditary and metabolic predisposition to keel deformity, the aetiology of keel deformity in modern hybrid layers seems more likely to involve external trauma rather than developmental defects. Old, healed, fractures of the furculum and keel appear to be more common in free-range and perchery birds than in caged layers (Gregory et al., 1990, 1994) and the cause for this increased prevalence is thought to be flight and landing accidents (Gregory and Wilkins, 1996). A recent comparison of floor and free-range systems with cages by Keutgen et al. (1999) revealed that the problem of keel bone ‘deviation’ or ‘deformation’ persists, particularly in the systems allowing increased activity. The European Union has recently passed legislation for the protection of egg laying hens that will alter husbandry conditions dramatically (Directive 1999/74/EC). This legislation encourages the use of more active systems to improve bird welfare. However, even in caged birds with no perches, severe keel deformities can occur, therefore increase in the use of more active systems may ironically lead to a greater risk of keel deformity. Gregory et al. (1991) noted that housing system can have a marked effect on keel fracture incidence. Percentages of birds with broken keels ranged from 2.7% (Elson terrace system) to16.0%(Gleadthorpepercherysystem).However, current data on the occurrence of keel deformities in layers are scarce. The object of the present study was to obtain data on keel deformities and relationships with other skeletal characteristics in hens of different strains kept under both commercial and experimental conditions. Effects of husbandry system were investigated by comparing birds kept under free-range and conventional cage systems. To investigate genetic effects, comparisons were made on birds divergently selected on the basis of skeletal characteristics (Bishop et al., 2000). Histological examinations were also performed throughout the entire study to establish the pathological nature of keels representative of different types of deformity.

MATERIALS AND METHODS Birds and husbandry Farm 1

Flocks of 28 000 chicks of Strain A (ISA) and 28 000 of Strain B (Hy-line) were floor reared until 16 weeks of age. They were then transferred either to battery cages (7 birds/cage, 630 cm2/ bird) or a litter and slat free-range system (11.7 birds/m2 ¼ 850 cm2/bird) until end of lay. From

321

these birds the following numbers were sampled evenly throughout the flock at 70 weeks and had keel, right tibia and humerus removed at culling for analyses: 100 of Strain A from cages, 50 of Strain B from cages, and 100 of Strain B from the free-range system. Farm 2

Chicks of Strain A (52 160) and chicks of Strain B (56 896) were reared in communal cages from one day old with no access to perches. At 16 weeks they were transferred to battery cages (5 or 10 birds/cage), each bird having 485 to 500 cm2 floor space. From these birds 75 of Strain A and 75 of Strain B were culled evenly throughout the flock at 70 weeks and had keel, tibia and humerus removed for analyses. Farm 3

Chicks of Strain A (660) were reared on a litter and wire system from one day old with no access to perches. At 15 weeks 132 hens were culled (as part of a separate study) and keels were examined radiographically. At 18 weeks the remainder were individually housed in battery cages, each hen having 450 cm2 of floor space. At 25 weeks a further 213 hens were culled for keel examination, and finally 285 surviving hens at end of lay (70 weeks) had keels, right tibiae and humeri assessed. Farm 4

Chicks of Strain C (625 LSL White Leghorn) were raised in brooders from one day old until 3 weeks and in communal rearing cages (192 60 46 cm, 10 birds per cage) until 15 weeks of age when they were transferred to individual battery cages (45 40 60 cm). The 605 surviving hens at end of lay (70 weeks) had keels, right tibiae and humeri assessed. Genetic study

A breeding programme utilising a restricted selection index to produce two lines of laying hens of Strain C with diverging skeletal characteristics has been described by Bishop et al. (2000). Chicks from generation 0 (G0, unselected), through to generation 7 (G7) of this study were reared in group cages (192 60 46 cm, 10 birds per cage) and transferred to individual battery cages (22 30 46 cm) at 16 weeks of age. They were given standard layer diets and water ad libitum until 68 weeks of age. Numbers of keels examined from birds in each generation are detailed in Table 3.

322

R.H. FLEMING ET AL.

Dissection protocol In all studies following culling, 4 cm of keel was sawn from the sternum along the base of the facies muscularis sterni from the apex sterni to include most of the corpus sterni (see Figure 1). The excised keels were examined for deformities and categorised in three groups: ‘normal’, ‘twisted’ (mainly folded but also compressions of the carina sterni) and ‘severe’ (see Figure 2). They were then fixed in 70% alcohol for

Pila carinae Apex sterni

A Corpus sterni

B

Linea intermuscularis sternalis

Carina sterni

Figure 1. Left lateral view of avian sternum, showing area A removed for radiographic examination (dotted lines) and region B used for ‘push-out’ shear strength test (dotted circle).

(a)

subsequent radiographic analysis. All long bones (both tarsometatarsi, tibiotarsi, femora, humeri, radii and ulnae) were examined during dissection for signs of fracture, fresh or healed, and the percentage of breakages of these bones was categorised as ‘other fractures’. The percentage of birds having two or more fractures (old or fresh) was also recorded. Radiographic density measurements X-rays of all keels were made in a Faxitron 804 soft X-ray unit (22 kV and 3 mA, 15 s exposure) using Kodak MRE-1 mammography film and Kodak Min-R2 cassettes with Min-R intensifying screen. This produced a high resolution radiograph as shown in Figure 3. Keels were removed from the 70% alcohol, blotted and placed on each 24 30 cm X-ray plate (30 keels per plate). A 16 step 0.25 mm increment aluminium stepwedge was also exposed on each plate at the same time as the keels for cross-calibration purposes. The resultant films were digitised on a voltage stabilised light-box via a Panasonic WV-BL600 monochrome CCD video camera (with the auto-gain setting OFF) linked to a

(b)

(c)

2cm Figure 2. Excised keels showing (a) normal appearance, (b) ‘twisted’ keel and (c) keel with a severe deformity.

(a)

(b)

(c)

(d)

(e)

Figure 3. Lateral X-rays of keels showing the radiographic appearance of (a) normal, (b) compressed, (c) twisted and (d) a severe deformity (with radio-dense fracture callus) in hens. For comparison, (e) shows a normal age-matched male keel with clearly defined trabeculae due to the absence of medullary bone.

323


(c)

(b) (a) (d)

Figure 4. Photograph of the keel ‘push-out’ shear testing device, showing (a) pressure transducer, (b) 6 mm push-out probe, (c) manual turnbuckle and (d) peak hold multimeter.

framegrabber in a Macintosh computer. A short macro routine was written for the public domain software package, NIH-Image v.1.62 (URL: http://rsb.info.nih.gov/nih-image/) which was used to display the keel image in false colour, and to calibrate each X-ray from the step-wedge image in terms of millimetres of aluminium equivalence (mm Al equiv.). This software could then delineate the boundary edge of each excised keel and calculate mean keel radiographic density (keel RD, in mm Al equiv.) for pixels contained within the boundary. Keels with pathological deformities attributed to traumatic injury were removed from the X-ray analysis as the presence of fracture callus material increased radiographic density significantly (see Figure 3d). Biomechanics Keel shear strength

A device based around a pressure transducer (Data Instruments, 0 to 50 mV) was manufactured from first principles (see Figure 4). The output from the transducer was amplified and recorded on a Metrix MX53 peak hold multimeter. Following dissection of the breast muscle around the keel, the shear force was applied manually via a turnbuckle across the bony plates of keel 5 cm from the proximal edge and 1 cm from the anterior edge (as marked in Figure 1). At the point of failure, a 6 mm disc of keel bone was pushed out by the probe and the peak mV reading recorded on the multimeter. This was converted to N/mm2 by laboratory calibration of the transducer from voltage to pressure, then to Newtons by correction for the probe crosssectional area. Tibiotarsus and humerus breaking strength

Following dissection of the right tibia and humerus of each hen, breaking strength by

three-point bending was determined on either an Instron materials testing machine with a connected plotter, or a JJ Lloyd LRX50 materials testing machine running the software package Nexygen 2.0. The bending jig and testing protocols for both machines were identical. The jig consisted of two 10 mm diameter steel bar supports, 30 mm apart at the centre, and a 10 mm diameter cross-head which approached at 30 mm/min. For both machines the breaking strength was determined from the peak of the load—displacement curve. Histopathology Keels representative of ‘twists’ and severe deformities, along with some of normal pathological appearance were decalcified in EDTA and processed for routine paraffin histology. Sections were cut at 6 mm transversely through each keel from the apex to the base and stained with haematoxylin and eosin. Sections were examined and histopathological evidence of healing bone was noted in each type of deformity. Collagen biochemistry of the pyrrolic cross-link Following preparation of tibiae and humeri samples from 117 hens in G6 of the genetic selection study as described by Sparke et al. (2002), the putative pyrrolic cross-link was assayed by reaction with p-dimethylaminobenzaldehyde, the colour values being calculated using a molar extinction coefficient based on that of N-methylpyrrole (Sims et al., 2000). Bone ash and mineral content The diaphysis of the left tibiotarsus, humerus and the entire keel from 117 hens in G6 of the genetic selection study was extracted in petroleum spirit, dried at 100 C and weighed. The

324

R.H. FLEMING ET AL.

bone was then heated in a muffle furnace at 550 C for 16 h and re-weighed, giving the percentage of total mineral content (ash) of the dry, fat-free sample. Genetic selection procedure Selection was performed retrospectively each year on the basis of data collected on skeletal characteristics from hens at the end of the laying period. Selection decisions were made with entire full sib families being kept or rejected. Preliminary statistical analyses on G1 and G2 data determined traits that were likely to be moderately or strongly inherited. A restricted selection index, designed to improve bone characteristics yet hold body weight (BW) constant, was derived from genetic variables obtained from these preliminary analyses using standard selection index theory (Bishop et al., 2000). Three biologically meaningful and moderately to highly heritable traits were included in the index: keel RD, humeral strength (HSTR) and tibial strength (TSTR). The resultant bone index (BI) was: BI ¼ (0.27 keel RDs) þ (0.37 HSTRs) þ (0.61 TSTRs) (0.25 BWs), where subscript ‘s’ implies that the trait has been standardised to have a standard deviation of 1.0 and a mean of 0. The coefficient for body weight was increased to 0.35 between G3 and G5 to counter a slight divergence in this trait appearing between the lines. Selection commenced from G3 onwards by assigning birds to either a high BI or low BI line on the basis of their dam’s BI from the previous generation. Statistical analyses Data were analysed by chi-squared analysis (keel fracture incidences, between strain, age, and housing type in the farm survey, and between BI in the genetic study) or by analysis of variance (ANOVA), for keel RD values and relationships

between keel deformity and body weights and strength of other bones. The software package Statview 5.0 (SAS Institute, Cary, NC; URL: www.statview.com) was used throughout. Fisher’s protected Least Significant Difference (PLSD) was used as the post hoc test for all possible pair-wise comparisons within categories or treatments.

RESULTS Farm survey Incidence of keel deformities

The incidences of keel deformities observed on each farm are shown in Table 1. Where possible, between-strain comparisons were made. Comparisons between strains at individual farms showed conflicting results, with Strain A hens having more normal keels than Strain B at Farm 1 (P ¼ 0.05), but more deformed keels at Farm 2 (P ¼ 0.04). Between-strain incidences of other fractures were not different at Farms 1 and 2. There was no significant difference between caged and free-range hens of Strain B at Farm 1, with proportions of normal, twisted and severely deformed keels being almost identical. Other fractures were not observed in free-range hens at Farm 1 whereas 3% of the same strain of caged hens showed other fractures. Table 2 shows the effect of age on incidence of keel deformities recorded at Farm 3. At 15 weeks of age, 99.2% of hens had normal keels, the remaining one hen (0.8%) had a slightly twisted keel. The percentage of hens with normal keels decreased by 25 weeks to 95.3% (P ¼ 0.09), the remaining hens having twisted keels, and this figure was further reduced from 25 weeks to 93.7% by 70 weeks (2 ¼ 3.7, P ¼ 0.05). Between-farm comparisons of incidence of keel deformity could not be made because, in addition to varying husbandry conditions outlined, birds were different ages at final culling.

Table 1. Occurrence of keel deformities and other fractures in a farm survey of hens of different strains Farm

No. birds Strain examined

Degree of keel deformity

Keel RD (mm Al equiv.)

% % % 2 Keel Normal Twists Severe status strain

Farm 1 Farm 1 (free range) Farm 2 Farm 3 Farm 4

100 50 100

A B B

94 83.3 84

5 14.6 14

1 2.1

75 74 285 605

A B A C

92 97.4 93.7 86.1

8 1.3 5.3 13.2

0 1.3

2

1 0.7

Percentage of hens with other fractures Old Peri-mortem Multiple Total % fracture fracture fractures with other fractures

6.02 P ¼ 0.05 —

0.55 0.01a 0.55 0.01a 0.67 0.01b

1 1 0

4 2 0

1 0 0

6 3 0

6.57 P ¼ 0.04 — —

0.70 0.01 0.74 0.01 0.67 0.01 0.59 0.005

0 0 2.1 2.5

1 0 2.1 0.5

0 0 0 0

1 0 4.2 3

Different superscripts denote significant within-farm differences in housing or strain.


Table 2. Incidence of keel deformities at different ages in hens from Farm 3 Age (weeks)

Number of keels examined

15 25 70

132 213 285

Keel status % Normal

% Twists

% Severe

99.2 95.3 93.7

0.8 4.7 5.3

0 0 1

Keel radiographic densities

There were no between-strain differences in keel RD values where comparisons were possible (at Farm 1 or 2) but there was a clear increase in keel RD values at Farm 1 in free-range hens compared to caged hens of the same strain (freerange ¼ 0.67 0.01, caged ¼ 0.55 0.01, P < 0.001). Genetic study

325

are significantly higher in high BI males for G4, G5 and G6, but the difference in G7 is not significant, possibly because of much lower sample numbers than previous generations (n ¼ 14 for both high and low BI males). Effect of genetic selection on keel push-out shear strength

Keel push-out shear strength measured in 218 high BI hens from G6 was significantly higher than keel shear strength in 197 low BI hens from G6 (60.2 1.2 N vs 56.6 1.2 N, P ¼ 0.03). Keel deformities and strengths of other bones

Hens from G3 to G7 inclusive with normal keels had significantly stronger tibiotarsus and humerus breaking strengths than those with deformities, confirming that keel deformity was associated with a weaker skeleton generally (see Table 5).

Incidence of keel deformities

The proportions of each type of keel deformity in G0 unselected birds, and selected birds up to G7 are shown in Table 3. Low BI hens had significantly more fresh fractures, twists and severe deformities than high BI hens for each generation, G3 to G7. For all generations (G3 to G7 pooled data), low BI hens had significantly more fresh fractures, twists and severe deformities than high BI hens. No keel deformities of any type were observed in male birds from either line. Between-generation comparisons were not made since generations were culled at various ages (68 to 72 weeks). Incidence of other fractures

Table 3 also shows the incidence of other (nonkeel) fractures. For all high and low BI G3 to G7 hens there were 281 birds with fresh or old fractures (21.7% of all hens). Of these birds, 4 (0.3% of all birds) had a fractured femur (two in conjunction with another fracture), two (0.15% of all birds) had a fractured ulna (with no other fracture present), one (0.08% of all birds) had a fractured tarsometatarsus (with other fractures) but the majority of these birds (278, or 21.5% of all birds) had breakages of one or both humeri. Radiographic densities

RD values from the genetic study are shown in Table 4. From G3 to G7, high BI hens had higher keel RD values than low BI hens (P < 0.0001 for all). Male keel RD values were measured from G4 to G7 and are also shown in Table 4. These

Keel deformities and body weights

Table 5 also shows the relationship between hens with keel deformities and body weight for all hens from G3 to G7. Hens with twisted keels are lighter than those with normal keels (twisted ¼ 1641.8 8.7 g, normal ¼ 1703.4 5.1 g, P < 0.0001). However, hens with severe keel deformities were of a similar body weight (1701.1 27.5 g) to those with normal keels. Keel deformities, collagen cross-links and bone ash in other bones

Table 6 shows no significant difference in measurements of bone quality (pyrrolic crosslinks or bone ash content in tibiotarsus and humerus), when split by type of keel deformity. Histopathology of keel deformities Examination of histological sections representative of different types of deformity revealed fracture callus of varying amounts and maturity (see Figure 5). In all cases fibro-cartilaginous callus appeared to have arisen from a periosteal source following a fracture along some part of the keel process. Fracture callus was found even in the less severe ‘twists’ of the keel (see Figure 5b). In deformities classed as ‘severe’, callus could occur along the base of the process, indicating that the entire corpus sterni had probably been moving with the action of the attached musculature. Severe deformities often involved displacement of large fragments of bone (see Figure 5d) with abundant bridging callus, as described by Hulth (1989). In some instances

326

R.H. FLEMING ET AL.

Table 3. Percentages of birds with different degrees of keel deformity and other fractures at end of lay in successive generations divergently selected for high or low bone index Generation

Unselected G3 Hens G4 Hens G4 Males G5 Hens G5 Males G6 Hens G6 Males G7 Hens G7 Males Overall % G3 to G7 Hens

No. birds

49 68 67 158 177 29 30 179 174 22 22 103 118 38 40 124 128 14 14 632 664

Line

High Low High Low High Low High Low High Low High Low High Low High Low High Low High Low

Degree of keel deformity % Normal

% Twists

% Severe

Keel status line

Old fracture

Peri-mortem fracture

Multiple fractures

Total % with other fractures

65.3 79.4 67.2 77.2 71.8 100 100 77.7 60.3 100 100 90.3 66.1 100 100 100 85.1 100 100 84.2 69.9

34.7 20.6 31.3 20.9 27.1 0 0 21.2 39.1 0 0 5.8 22.9 0 0 0 14.1 0 0 14.4 27.4

0 0 1.5 1.9 1.1 0 0 1.1 0.6 0 0 3.9 11.0 0 0 0 0.8 0 0 1.4 2.7

— 4.91 NS 1.25 NS

4.1 0 0 3.2 2.8 0 0 0 2.9 0 0 0.98 2.5 0 0 0 0.8 0 0 0.9 2.1

10.2 14.7 20.9 19.6 24.9 0 0 2.8 16.1 0 0 15.5 40.7 0 0 0 14.8 0 0 9.8 23.1

6.1 2.9 7.5 3.8 4.5 0 0 0.6 2.3 0 0 2.9 11.0 0 0 0 3.1 0 0 1.9 5.1

20.4 17.6 28.4 26.6 32.2 0 0 3.4 21.3 0 0 19.4 54.2 0 0 0 18.7 0 0 12.6 30.3

Table 4. Keel radiographic density (RD) values in birds at end of lay in successive generations divergently selected for high or low bone index Generation

Unselected hens G3 Hens G4 Hens G4 Males G5 Hens G5 Males G6 Hens G6 Males G7 Hens G7 Males

Percentage of birds with other fractures 2

No. birds

Line

Keel RD (mm Al equiv.)

146 68 67 158 177 29 30 179 174 22 22 103 118 38 40 124 128 14 14

— High Low High Low High Low High Low High Low High Low High Low High Low High Low

0.43 0.01 0.57 0.01a 0.50 0.01b 0.52 0.01a 0.47 0.01b 0.68 0.01a 0.57 0.01b 0.58 0.01a 0.48 0.01b 0.77 0.01a 0.67 0.02b 0.53 0.01a 0.46 0.01b 0.70 0.01a 0.62 0.01b 0.51 0.01a 0.45 0.01b 0.50 0.01 0.47 0.01

Values with different superscripts within each generation and gender are significantly different.

the appearance of woven fracture callus was indistinguishable from the appearance of medullary bone (see Figure 5g), except of course, that areas of fracture callus were external to the periosteal surface. All deformities examined appeared to have been caused by traumatic fracture rather than any congenital or developmental defect.

NS 7.75 P ¼ 0.02 NS 16.92 P ¼ 0.0002 NS 16.22 P ¼ 0.0003 NS 6.39 P ¼ 0.04

Table 5. Bone strengths of other bones and body weights of hens from G3 to G7 in the genetic study with different degrees of keel deformity Degree of keel deformity Normal Twisted Severe

Breaking strength (N SEM) n

Tibiotarsus

Humerus

134.2 1.2a 1027 257.0 2.8a 287 215.2 3.9b 120.4 2.1b 26 213.4 14.1b 111.3 7.3b

Body weight (g) 1703.4 5.1a 1641.8 8.7b 1701.1 27.5a

Values with different superscripts in columns are significantly different.

Table 6. Pyrrolic cross-link and bone ash content of tibiotarsus and humerus in a sub-sample of hens from G6 of the genetic study with different degrees of keel deformity Degree of Pyrrolic cross-link content Bone ash content (%) keel (mole/mole collagen SEM) deformity n Tibiotarsus Humerus Tibiotarsus Humerus Normal Twisted Severe

119 0.46 0.02 0.41 0.34 56.9 0.4 69.9 0.2 24 0.49 0.03 0.30 0.03 56.9 0.7 69.3 0.4 3 0.46 0.04 0.42 0.08 59.5 1.6 71.1 0.4

DISCUSSION The incidence of keel deformities reported here in caged laying birds from various flocks ranges from 2.6% ‘twisted’ and ‘severe’ deformities combined (Farm 2, Strain B) to 16.7% (Farm 1, Strain B). Extrapolation of these figures suggests that of a current EU flock of around 300 million caged laying hens (Windhorst, 2001), at best around 8 million, and at worst, around 50 million

327


5mm

D

C

A

B

(e)

MB

FC

NB

Co.

NB 100µm

200µm

(f)

(g)

MB

100µm (h)

Figure 5. Photomicrographs of hen keels sectioned 1 cm distal to the apex sterni, through the corpus sterni show (a) normal histological appearance, (b) appearance of a ‘twisted’ keel, (c) and (d) appearance of keels with deformities classed as ‘severe’. In (d) the leading edge of the keel is poorly reduced and movement (arrowed) has produced abundant fracture callus (FC). View (e) shows increased magnification of area A, ( f ) shows area B, with fibro-cartilaginous new bone (NB), (g) shows detail of area C with similarities in the appearance of woven medullary bone (MB) and FC, separated by the original cortical bone (Co.). View (h) is a detail of area D, again with new bone, alongside darker MB (H&E stain throughout).

hens are likely to be suffering some kind of deformity of the keel bone, up to 6.3 million (2.1%) of them with what can be classed as a severe keel deformity. These deformities, whether slight or severe, represent a chronic welfare problem in laying hens. Examination of keels from hens at different ages (15, 25 and 70 weeks) from Farm 3 revealed that keel deformity was not evident in pullets prior to lay, but incidence of deformities appeared to increase during the laying period. This is in

accordance with previous observations that measures of skeletal strength and structure deteriorate during the laying period in hens due to the onset of osteoporosis (Fleming et al., 1998) and this suggests that keel deformity is due to the same loss of bone with age. The present study may provide a baseline for future comparisons of keel condition in hens from more active husbandry systems. Over the last decade various studies have revealed that husbandry systems that promote

328

R.H. FLEMING ET AL.

activity are beneficial for laying hens in terms of skeletal mass (Knowles and Broom, 1990; Fleming et al., 1994; Newman and Leeson, 1998). In a study of 7 alternative cage types, a simple increase in cage height was shown to improve strength of the humerus (Moinard et al., 1998). In the present study keel condition was examined in 100 hens from a commercial free-range system (Farm 1, Strain B, free-range). Keels were in similar condition to those from caged siblings at the same farm. In addition, there were no other fractures in other bones examined in these 100 hens and keel RD values were greater than caged equivalents (P < 0.001), suggesting that overall skeletal condition was improved by removal of the immobilisation effect of cages. It should be noted that hens in the free-range system were only afforded an extra 220 cm2 per bird more than the 630 cm2 per bird in the cages, however, the increased freedom to move around, with outdoor access, improved skeletal condition. This skeletal improvement was probably enough to overcome any detrimental effects (such as increased collision risk) of a system allowing increased activity. However, our results are not in agreement with other surveys of skeletal condition in hens from alternative systems (NorgaardNielsen, 1990; Gregory et al., 1994; Gregory and Wilkins, 1996), where more active systems fared badly in terms of old, healing fractures. Keutgen et al. (1999) used a different scoring system to the present study, therefore direct comparisons are difficult, but keel bone deformation was reported to be more than doubled in free-range and floor systems compared to caged birds. This information on incidence of keel deformities in hens is timely as the European Union has recently passed directives for the protection of egg laying hens that will alter husbandry conditions dramatically (Directive 1999/74/EC). These changes will introduce alternative conditions for layers in the form of aviary, perchery or free-range systems, or as larger, communal ‘enriched’ cages with perches, litter and nesting areas. The introduction and design of perches in aviary systems and enriched cages raises some problematic issues. Levels of keel deformity observed in relatively inactive caged birds in our farm survey are already alarmingly high. In the more open aviary systems or in enriched cages incorporating perches, damage may be even more likely to arise through violent contact between keel and perch following poor landings although this was not found in the present trial. Appleby (1993), and Gregory and Wilkins (1996) have noted that keel bone deformation is associated with the presence of perches. Studies by Tauson and Abrahamsson (1994, 1996) suggested that a hardwood circular perch of 38 mm diameter with flattened upper

and lower surfaces reduced keel bone lesions in get-away cages compared with perches of rectangular cross section. Future studies should investigate the effect of perch composition upon keel integrity, and indeed the design and layout of all hard surfaces and landing areas in alternative systems to avoid unnecessary welfare problems. The genetic study has shown that genetic selection can improve keel condition. The general skeletal condition of the high BI hens improved compared to low BI hens in each successive generation. In addition, G7 high BI hens had no keel deformities and no other fractures in the bones examined compared to 14.9% of hens with deformed keels and 18.7% of hens having other fractures in the low BI line. No keel deformity was observed in any male bird, and male keel RD values were always higher than hen keel RDs in all generations. Interestingly, high BI males had keels that were significantly more radio-dense than those of low BI males in G4, G5 and G6, even although selection decisions were made entirely on the basis of the BI of the dams alone, as described by Bishop et al. (2000). Fracture incidences in the genetic study compared to the farm survey were higher and the likely explanation for this is the method used at culling; in the farm survey, birds were culled by overdose of barbiturate, whilst in the genetic survey, birds were culled throughout by cervical dislocation, causing wing flapping that probably resulted in more peri-mortem fractures of the humerus in particular. Results from the genetic study also established that hens with normal keels had significantly stronger bones in the appendicular skeleton (tibiotarsus and humerus) than hens with deformed keels. Qualitative assessments of other bones in hens from the genetic study with keel deformities show no differences in the variables studied (bone ash and pyrrolic cross-links). We conclude from this, and the reduction in strength of appendicular bones in hens with keel deformities, that lack of bone mass is the cause of keel fracture and deformity rather than the quality of the bone present. No simple relationship between body weight and the presence of a keel deformity was found. In a study of fracture incidence by Knowles and Broom (1993), tibiotarsal strength increased with body weight but the probability of a bone being broken also increased with weight, the overall effect being that the increase in bone strength was not great enough to prevent the additional damage suffered by heavier hens. However, the keel is not a load bearing bone, and a clear relationship between likelihood of fracture and body weight does not seem to exist in this case.

329


The histology of deformed keels in this survey reveals that fracture callus is present in varying amounts and stages. If these deformities are due to fracture, they represent a problem for welfare in terms of ‘freedom from pain’. Studies in fracture repair demonstrate that generally, the greater the amount of callus found in a healing fracture, the greater the amount of movement occurring during the healing process (Hulth, 1989). The acute response to nociceptive stimuli in the domestic hen is similar to that in mammals, and the perception of chronic pain during the healing period following trauma is also similar (Gentle, 1992). Chronic pain perception in particular presents a major welfare problem. The periosteal layer covering bones contains ‘abundant myelinated and nonmyelinated nerve fibres ending in networks on the surfaces of bone tissue’ (McLean and Urist, 1968). Therefore, unstable fractures in bones ensheathed with this neuroreceptor-rich periosteal layer, such as the keel, will undoubtedly cause pain at fracture and during the course of healing. These fractures may often be poorly reduced and mobile, with the action of the chest musculature causing additional movement and discomfort. A fractured keel is unlikely to be detected in a commercial laying hen house as easily as a long bone fracture and the animal may experience prolonged unnecessary suffering as a result of this. In conclusion, we have demonstrated standard selection techniques based on familial traits can improve keel quality significantly in laying hens. Good keel quality should be a prerequisite for these new systems since the keel appears particularly vulnerable to fracture, as evidenced in our survey of commercial and research-based poultry. In addition, we have demonstrated that hens with keel deformities are more likely to be those with weaker bones overall and these particular animals may be more likely to break other bones in husbandry systems allowing higher activity. Our results show that current levels of keel deformity in layers can be high, even in cages without perches, and raise some concerns about the imminent introduction of higher activity husbandry systems where more landing accidents on perches and other hard surfaces may occur. We suggest that genetic selection should be used as a means of improving the skeletal characteristics of hens so that keel deformities and other skeletal damage can be minimised in future husbandry systems.

ACKNOWLEDGEMENTS The authors wish to thank those farms participating in the survey, and Lohmann Tierzucht GmbH (Cuxhaven, Germany) for assistance with

the genetic study. We also thank the students and staff of the Department of Mechanical Engineering, University of Edinburgh, for the design and build of the keel shear-testing device. The Muscle and Collagen Group, University of Bristol provided the cross-linkage data. We wish to thank DEFRA and the EC for funding these studies.

REFERENCES APPLEBY, M.C. (1993) Should cages for laying hens be banned or modified? Animal Welfare, 2: 67—80. BISHOP, S.C., FLEMING, R.H., MCCORMACK, H.A., FLOCK, D.K. & WHITEHEAD, C.C. (2000) Inheritance of bone characteristics affecting osteoporosis in laying hens. British Poultry Science, 41: 33—40. BUCKNER, G.D., INSKO, W.M. & HENRY, A.H. (1946) Deformed bones of chickens. I. Crooked keel. Bulletin of the Kentucky Agricultural Experimental Station, 492: 1—14. EC COUNCIL DIRECTIVE 1999/74/EC Laying down minimum standards for the protection of laying hens. Official Journal of the European Communities, L203: 53—57. FLEMING, R.H., MCCORMACK, H.A. & WHITEHEAD, C.C. (1998) Bone structure and strength at different ages in laying hens and effects of dietary particulate limestone, vitamin K and ascorbic acid. British Poultry Science, 39: 434—440. FLEMING, R.H., WHITEHEAD, C.C., ALVEY, D., GREGORY, N.G. & WILKINS, L.J. (1994) Bone structure and breaking strength in laying hens housed in different husbandry systems. British Poultry Science, 35: 651—662. GENTLE, M.J. (1992) Pain in birds. Animal Welfare, 1: 235—247. GREGORY, N.G. & DEVINE, C.D. (1999) Body condition in end-of-lay hens: some implications. Veterinary Record, 145: 49. GREGORY, N.G. & WILKINS, L.J. (1989) Broken bones in domestic fowl: handling and processing damage in end-of-lay battery hens. British Poultry Science, 30: 555—562. GREGORY, N.G. & WILKINS, L.J. (1996) Effect of age on bone strength and the prevalence of broken bones in perchery laying hens. New Zealand Veterinary Journal, 44(1): 31—32. GREGORY, N.G., WILKINS, L.J., ELEPERUMA, S.D., BALLANTYNE, A.J. & OVERFIELD, N.D. (1990) Broken bones in domestic fowl: effect of husbandry system and stunning method in end-of-lay hens. British Poultry Science, 31: 59—69. GREGORY, N.G., WILKINS, L.J., KESTIN, S.C., BELYAVIN, C.G. & ALVEY, D.M. (1991) Effect of husbandry system on broken bones and bone strength in hens. Veterinary Record, 128: 397—399. GREGORY, N.G., WILKINS, L.J., KNOWLES, T.G., SORENSEN, P. & VAN NIEKERK, T. (1994) Incidence of bone fractures in European layers. Proceedings of the 9th European Poultry Conference, Glasgow, pp. 126—128. HULTH, A. (1989) Current concepts of fracture healing. Clinical Orthopaedics and Related Research, 249: 265—283. KEUTGEN, H., WURM, S. & UEBERSCHAR, S. (1999) Pathologisch-anatomische Untersuchungen bei Legehennen aus verschiedenen Haltungssystmen. Deutsche Tierartzliche Wochenschrift, 106: 125—188. KNOWLES, T.G. & BROOM, D.M. (1990) Limb bone strength and movement in laying hens from different housing systems. Veterinary Record, 126: 354—356. KNOWLES, T.G. & BROOM, D.M. (1993) Effect of bone strength on the frequency of broken bones in hens. Research in Veterinary Science, 54: 15—19.

330

R.H. FLEMING ET AL.

MCLEAN, F.C. & URIST, M.R. (1968) Bone: Fundamentals of the Physiology of Skeletal Tissue, p. 17 (Chicago, University of Chicago Press). MOINARD, C., MORISSE, J.P. & FAURE, J.M. (1998) Effect of cage area, cage height and perches on feather condition, bone breakage and mortality of laying hens. British Poultry Science, 39: 198—202. NEWMAN, S.C. & LEESON, S. (1998) Effect of housing birds in cages or an aviary system on bone characteristics. Poultry Science, 77: 1492—1496. NORGAARD-NIELSEN, G. (1990) Bone strength of laying hens kept in an alternative system, compared with hens in cages and on deep-litter. British Poultry Science, 31: 81—89. SIMS, T.J., AVERY, N.C. & BAILEY, A.J. (2000) Quantitative determination of collagen crosslinks, in: STEULI, C. & GRANT, M.E. (Eds) Methods in Molecular Biology, Vol. 139, Extracellular Matrix Protocols (Totowa, NJ, Humana Press). SPARKE, A.J., SIMS, T.J., AVERY, N.C., BAILEY, A.J., FLEMING, R.H. & WHITEHEAD, C.C. (2002) Differences in

composition of avian bone collagen following genetic selection for resistance to osteoporosis. British Poultry Science, 43: 127—134. TAUSON, R. & ABRAHAMSSON, P. (1994) Foot and skeletal disorders in laying hens; effects of perch design, hybrid, housing system and stocking density. Acta agriculturae Scandinavica, 44: 110—119. TAUSON, R. & ABRAHAMSSON, P. (1996) Foot and keel bone disorders in laying hens; effects of artificial perch material and hybrid. Acta agriculturae Scandinavica, 46: 239. WARREN, D.E. (1937) Physiologic and genetic studies of crooked keels in chickens. Kansas Agricultural Experimental Station Technical Bulletin, 44. WHITEHEAD, C.C. & FLEMING, R.H. (2000) Osteoporosis in cage layers. Poultry Science, 79: 1033—1041. WINDHORST, H.-W. (2001) Is there a future for Europe’s egg industry? Poultry International, 40(7): 26—33. WINTER, A.R. & FUNK, E.M. (1941) Poultry Science and Practice, p. 388 (Chicago, J.B. Lippincott).