Development and evaluation of two different lameness models in meat goats, a pilot study

Abstract

Lameness is a serious health concern for livestock species. Understanding individual animal response to pain and characterization of lameness are critical when developing appropriate treatments. The objectives of this pilot study was to evaluate two different lameness models and measures for determining response to induced lameness in meat goats. Lameness was induced by intraarticular injection into the left hind lateral claw distal interphalangeal joint with either amphotericin B (Amp-B) or kaolin-carrageenan (K-C). Response to lameness was characterized by behavior scoring, visual lameness scoring (VLS), infrared thermography (IRT) of the affected digit, pressure mat gait analysis (PMT), and plasma cortisol (CORT) analysis. Lame goats had higher VLS compared to controls (P = 0.003). Maximum temperatures measured in hooves from lame vs control goats were significantly higher (P = 0.003). Pressure mat analysis demonstrated, when compared to controls, lame goats had decreased force (P = 0.013), impulse (P = 0.007), contact pressure (P = 0.007), and contact area of the left hind limb (P = 0.009). Mean CORT levels 4 and 6 h after lameness induction were higher in lame goats (P = 0.005, P = 0.01). The two lameness methods reliably induced lameness of varying severity in healthy meat goats.

INTRODUCTION

Lameness is an important behavioral indicator of pain, affecting up to 67% of dairy goats (Hill et al., 1997; Christodoulopoulos, 2009; Crosby-Durrani et al., 2016). It is regarded as one of the most serious health and welfare concerns in ruminants (Christodoulopoulos, 2009). Hoof lesions associated with lameness in goats include sole ulcers, subsolar abscesses, and interdigital dermatitis (Hill et al., 1997; Christodoulopoulos, 2009; Crosby-Durrani et al., 2016). Such lesions, when left untreated, can result in deep digital sepsis and septic arthritis. The economic impact of lame animals is reflected by decreased productivity, cost of treatment and premature culling of animals (Christodoulopoulos, 2009). Welfare and economic concerns associated with lameness can be minimized if veterinarians can provide evidence-based treatment and prevention recommendations. Identification of lameness needs to be sensitive and reliable and therapeutic interventions need to be assessed using objective, validated methods to quantify lameness. Lameness models are the first and necessary step to objectively evaluate animal response to lameness. Intraarticular injection amphotericin B or kaolin-carrageenan are two different lameness models used in other species (McIlwraith et al., 1979; Coetzee et al., 2014, Muley et al., 2016).

Amphotericin B is a polyene antimicrobial that has been used to induce an acute, transient synovitis (McIlwraith et al., 1979; Coetzee et al., 2014). Intra-articular injection of Amp-B causes synovitis by destruction of lysosomes and release of inflammatory mediators (McIlwraith et al., 1979). The onset and severity of Amp-B synovitis differs greatly between species. In horses, Amp-B induces a moderate to severe lameness with a duration of 3 d to 2 wk (McIlwraith et al., 1979). A consistent, moderate lameness, peaking in severity from 6 to 12 h after injection, and lasting 72 h was produced in cattle (Schulz et al., 2011). In both species, the duration and severity of lameness induced was adequate for development of a validated lameness model. The advantages of this model are the moderate severity and short duration of lameness.

Kaolin-carrageenan is a widely used acute arthritis model producing a characteristic pattern of lameness in primates, mice, rats, and cats (Muley et al., 2016). Kaolin is an inorganic hydrated aluminum silicate. Intra articular injection of kaolin results in a physical irritation of the articular cartilage and synovium. Carrageenan is a sulfated polysaccharide that induces an acute lameness through a series of pro-inflammatory reactions terminating in leucocyte activation and pro-inflammatory cytokine release (Muley et al., 2016). The advantage of this lameness model is the well-defined onset of a robust inflammatory pattern resulting in neurogenic inflammation allowing for a more complete analysis of pain (Muley et al., 2016). The pain response is generally of short duration lasting no more than 24 h. We hypothesized that both Amp-B, and K-C would induce lameness, but the lameness induced using Amp-B would be less severe than lameness induced by K-C. To date, there have been no published studies evaluating lameness models or response to induced lameness in meat type goats. According to the most recent USDA NAHMS Goat 2009 study, the meat goat industry is one of the fastest growing segments in the U.S. livestock industry (USDA, 2010). In that study, Boer goats were one of the most popular breeds used for meat production (USDA, 2010). Due to the increasing popularity of the Boer breed, Boer goats were used in this study. The purpose of this pilot study was to develop two different lameness models using Amp-B or K-C and to evaluate measures for determining response to induced lameness in goats. The following specific outcome variables were measured in response to induced lameness; behavior, visual lameness scoring (VLS), infrared thermography (IRT) of the affected digit, pressure mat gait analysis (PMT), and plasma cortisol (CORT) analysis.

MATERIALS AND METHODS

This study was approved by the Institutional Animal Care and Use Committee at Kansas State University (Protocol #4178).

Animals and Study Design

Eighteen, approximately 8 mo of age (BW = 41 ± 20 kg), male intact Boer goats were used in this study. Prior to enrollment, all animals were determined to be sound and free of lameness by observation of gait as they walked in an alley, by an experienced veterinarian trained in visual lameness scoring of goats. Goats were housed individually in raised metal pens (1.5 × 2.4 m) with a grated floor and fed a custom mixed complete pelleted ration for the duration of the study. The goats had been handled and housed in the same individual pens during a feeding trial that ended 1 mo prior to the lameness study. Stride length was used to determine the number of goats needed per treatment group. The study was designed to have a statistical power of 0.80 assuming an effect size in differences of 10 cm and standard deviation of 8.5 cm. A sample size of six animals per group was determined based on the estimates of true difference and variability of measured outcome variables. A random number generator (random.org) was utilized to assign each of the goats to one of three treatment groups: control (CNTL), Amp-B, or K-C. The study occurred over a 3-wk period of time: wk 1 animals were trained to walk across the pressure mat by walking back and forth through the pressure mat testing alleyway to establish a consistent gait and reduce variation during the trial, wk 2 lameness was induced, wk 3 animals were monitored for any residual lameness.

Lameness Induction

Lameness induction occurred over 2 d. Nine goats were randomly selected to have lameness induced on d 1 and 9 goats had lameness induced on d 2. The right jugular vein of each goat was aseptically prepared and a 16-g × 7.5 cm catheter (Mila International, Inc., Florence, KY, USA) was steriled, placed to facilitate sedation, and blood sample collection. Each animal was heavily sedated (0.1 mg/kg) using xylazine hydrochloride 20 mg/mL (Akron, Inc., Lake Forest, IL, USA) administered intravenously through the catheter. Once sedated, each animal was placed in right lateral recumbency; and the hair over the coronary band of the left hind limb lateral claw was clipped and skin aseptically prepared. A veterinarian skilled in intraarticular injections, injected the distal interphalangeal joint of the left hind lateral claw with 1 mL of sterile saline 0.9% CNTL, 1 mL of Amp-B, or 1 mL K-C using an 18-g × 2.54 cm hypodermic needle (Exelint International, CO., Redondo Beach, CA, USA). Solutions were prepared as described below. Group CNTL (n = 6): a commercially prepared sterile 0.9% saline solution for injection (Baxter Healthcare Corporation, Deerfield, IL, USA) was used for intraarticular injection. Group Amp-B (n = 6): a sterile solution containing 10 mg/mL of amphotericin B was prepared by rapidly expressing 5 mL of sterile water for injection USP without a bacteriostatic agent directly into the 50 mg of amphotericin lyophilized cake (X-gen Pharmaceuticals, Inc., Big Flats, NY, USA), using a sterile needle and syringe. The contents were mixed immediately until the colloidal solution was clear. Group K-C (n = 6): a sterile solution containing 5% carrageenan plus 5% kaolin was prepared using non-gelling carrageenan and kaolin powder (Sigma-Aldrich Co., St. Louis, MO). The solution was shaken immediately prior to intraarticular administration.

Intraarticular administration was confirmed by lack of resistance to injection. If extreme resistance against the plunger of the syringe during injection occurred, the substance was considered to be administered periarticular and was noted in the record (n = 2). To disseminate the product within the distal interphalangeal joint, after injection, the joint was flexed and extended five times. Sedation was reversed in each goat using a single intramuscular injection (0.5 mg/kg) of tolazoline 100 mg/mL (Tolazil MWI, Boise, ID, USA). Goats were monitored hourly for behavioral signs of excessive pain over a period of 10 h following lameness induction followed by twice-daily monitoring for 7 d. Goats exhibiting a VLS >2 for >48 h, prolonged recumbency (recumbent for >1 h without rising when encouraged to stand), anorexia, or signs of depression received rescue analgesia with flunixin meglumine (BanamineMerck Animal Health Intervet Inc., Madison, NJ, USA) 2.2 mg/kg intravenously once every 24 h as needed (Amp-B n = 3, K-C n = 3).

Outcome Variables

Data parameters collected included: behavior scoring, VLS, IRT, PMT, and CORT. Data was collected at baseline, lameness induction, 4, 6, 12, 24, 48, and 72 h postlameness induction.

Behavior Scoring

Video cameras (Sony Handycam HDR-CX405, Sony USA Inc., NY, USA) were placed on tripods outside of each pen. Goats were video recorded the day prior to lameness induction at three times (0700, 1100, 1300 hours) for 10 min. Postlameness induction, goats were video recorded for 10 min at the following time points: 4, 6, 12, 24, 48, 72 h. The videos were randomized across goat treatment and time point using a random number generator (random.org) prior to being scored. Each goat was behavior scored continuously by a trained observer blinded to treatment and time point using BORIS software (Behavioral Observation Research Interactive Software v 7.7.3, Torino, Italy) and a detailed ethogram adapted from Walker et al. (2015; Table 1). The total duration of behaviors was converted into proportions of time prior to analysis to create time budgets.

Table 1.

Ethogram used to score goat behavior, grouped into maintenance, locomotion, oral behavior, social interaction and pain behavior (adapted from Walker et al., 2015)

Behaviors	Description
Eating	Ingesting food provided at feed bunk
Drinking	Consuming water from nozzle
Defecating	Passing fecal matter in standing or lying position
Urinating	Passing urine in standing or lying position
Sleeping	Lying down, eyes closed
Scratching	Using horns or rear hoof to scratch the body
Ruminating	Regurgitating, chewing, and swallowing food
Pawing	Moving front limb in a “digging” motion against the ground or feed bunk
Grooming	Licking or rubbing body or head against pen
Walking	Moving forward at a normal pace
Limping	Walking with one or more legs not supporting body weight
Standing	Body weight supported by four legs, no forward movement
Standing on three legs	Body weight supported by three legs; non-weight bearing on one limb. No forward movement
Sitting	Body weight supported by hindquarters and front legs
Lying	Recumbent, body on ground
Rearing	Body weight supported by back legs. Front legs on fence or feed bunk
Kneeling	Body weight supported by front carpal joints and hind legs
Licking	Moving tongue over surfaces or pen mates
Chewing	Nibbling at substrates or conspecific in nearby pen
Sniffing	Inhaling air close to object or conspecific in nearby pen
Playing	Running, trotting, galloping, or springing
Butting	Head-to-head or head-to-body contact with conspecific in nearby pen
Allo-grooming	Licking or rubbing body against conspecific in nearby pen
Restlessness	Repeated sitting, standing or walking for short durations, unsettled
Attention to hoof	Guarding or constant attention to injured foot or limb
Tail wagging	Tail movement from side to side (or up and down)

Visual Lameness Scoring

Using a previously described VLS (Deeming et al., 2018), a skilled observer (veterinarian with extensive clinical experience in identification of lameness in all food animals species) blinded to treatment scored each goat prior to lameness induction and at each sampling period postlameness induction (4, 6, 12, 24, 48, and 72 h) as the animals walked through an open-sided alley system with flat, non-slip flooring on their way to the pressure mat. Lameness was graded on a scale from 1 to 5 with the following grade specifications: 1—normal gait, 2—uneven gait, 3—mildly lame, 4—moderately lame, 5—severely lame.

Infrared Thermography

Three images of the dorso-lateral aspect of the affected foot (left hind) and contralateral unaffected foot were taken at each time point (baseline, lameness induction, 4, 6, 12, 24, 48, and 72 h postlameness induction) using a research grade infrared camera (FLUKE Ti580 IR Imager, Fluke Corp., Everett, WA, USA). Images were obtained at a 45° angle and 1 m from the coronary band. Infrared images were analyzed using research grade computer software (SmartView 4.3, FLUKE Corp., Everett, WA, USA) to determine maximum temperatures of the foot at the level of the coronary band.

Pressure Mat Gait Analysis

A commercially available floor mat-based pressure/force measurement system (Strideway, Tekscan, Inc., South Boston, MA, USA) was used to record and analyze the steps of each goat. Goats walked across the pressure mat system one time prior to lameness induction and one time at 4, 6, 12, 24, 48, and 72 h postlameness induction. Video synchronization was used to ensure consistent gait between and within goats. Using research grade software (Strideway v 7.7, Tekscan, Inc., South Boston, MA, USA), stance time, stride length, contact force, impulse, contact pressure, and contact area were assessed using the methods as described by Coetzee et al. (2014).

Plasma Cortisol

Blood was collected from the intravenous catheter prior to and at 4, 6, 12, 24, 48, and 72 h postlameness induction. Plasma cortisol concentrations were determined in duplicate via a radioimmune assay (MP biomedical Cortisol Coated Tube (CT) RIA kit catalog #: 07-221105R; MP Biomedical, Santa Ana, CA, USA) using methods adapted from Kleinhenz et al. (2017). The assay had a detection range of 0.64 to 150 ng/ml. The coefficient of variation for the intra-assay variability was 11.1% and the interassay variability was calculated to be 10.3%.

Statistical Analysis

Behavioral data was analyzed using a GLIMMIX procedure with a beta distribution, including time, treatment and the time × treatment interaction in SAS (Statistical Analysis System, v 9.4, SAS Institute Inc., NC, USA). Time was a repeated measure in the model, with goat as the experimental unit.

Plasma cortisol were log transformed for normality prior to statistical analysis. The outcome responses of IRT, PMT, and CORT were analyzed using linear mixed models with goat as the experimental unit. Goats nested in a treatment group were designated as a random effect, with treatment, time and treatment × time interaction designated as fixed effects.

Visual lameness scores were treated as categorical data and Fisher’s exact test was used to compare scores. All statistics, except for behavior data, were performed using statistical software (JMP Pro 14.0, SAS Institute, Cary, NC, USA). Post hoc tests were conducted on significant factors using the Tukey-Kramer adjustment. Statistical significance was set at P ≤ 0.05.

RESULTS

Behavior Score

The behavior results are presented in Table 2. Amphotericin B and K-C goats displayed similar levels of limping and standing on three legs (nonweight bearing on one limb) after lameness was induced. Goats in the CNTL group did not display either of these behaviors across the observation period. Amphotericin B and K-C goats urinated more than CNTL goats and K-C goats wagged their tails significantly more than CNTL goats. There were no significant differences in behavior between any of the treatment groups prelameness induction.

Table 2.

Proportion of time (mean ± SE) goats exhibited each behavior pre- and postlameness induction using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6)

		Prelameness induction		Postlameness Induction
	Behavior	Treatment P-value	Pretreatment	Treatment P-value	Time  P-value	Amp-B	K-C	CNTL
Proportion (duration)	Grooming	0.2707	0.07 ± 0.00	0.3489	0.0012	0.06 ± 0.01	0.06 ± 0.01	0.07 ± 0.01
	Scratching	0.7013	0.04 ± 0.00	0.8103	0.0007	0.04 ± 0.00	0.04 ± 0.00	0.03 ± 0.00
	Rearing	0.4830	0.27 ± 0.06	0.2626	0.0484	0.14 ± 0.05	0.11 ± 0.04	0.22 ± 0.04
	Tail wagging	0.3153	0.02 ± 0.00	0.0461	0.4901	0.01 ± 0.00ab	0.02 ± 0.00b	0.00 ± 0.00a
	Limping	–	– d	0.2159	0.2114	0.07 ± 0.02a	0.04 ± 0.00a	– b
	Standing (3 legs)	–	–	0.2500	0.5656	0.40 ± 0.09a	0.50 ± 0.07a	– b
	Urinating	–	0.04 ± 0.03	0.0326	–	0.23 ± 0.16a	0.14 ± 0.00a	0.04 ± 0.00b

^a,bMeans with different superscripts in the same row differ significantly (P < 0.05); entries in italic font are statistically significant.

^cOnly behavior variables that were significant posttreatment are presented.

^dDash indicates behavior was not observed.

Visual Lameness Score

One goat from each of the Amp-B and K-C groups failed to show clinical lameness after induction. After lameness induction there was a higher number of goats with a VLS of 2 or greater in the Amp-B and K-C groups (P = 0.003) at all time points (Figure 1).

Figure 1.

Visual lameness score counts for goats with induced lameness using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6).

Infrared Thermography (IRT)

There was evidence of treatment (P = 0.03), time (P < 0.001), and treatment × time (P = 0.11) interactions for maximum foot temperature taken by IRT (Figure 2; Table 3). Maximum temperature peaked at 12 h for all groups (Amp-B, CNTL, K-C) and were 37.65 °C (95% CI: 35.52 to 39.78 °C), 36.62 °C (95% CI: 34.49 to 38.74 °C), and 37.85 ± 1.11 °C (95% CI: 35.72 to 39.98 °C), respectively. Temperature of CNTL hooves were lower than those of the Amp-B and K-C groups at 24, 48, and 72 h postinduction. At 24 h postinduction, the maximum temperature of the CNTL goats’ hooves were 31.62 °C (95% CI: 29.49 to 33.74 °C) (P = 0.003) compared to Amp-B 36.00 °C (95% CI: 33.73 to 38.25 °C) and to K-C 36.67 °C (95% CI: 34.41 to 38.93 °C) goats. At 48 h postinduction, the maximum temperature of the CNTL goats’ hooves were 30.07 °C (95% CI: 27.94 to 32.19 °C) (P = 0.01) compared to the Amp-B and K-C goats at 34.43 °C (95% CI: 32.31 to 36.56 °C) and 33.88 °C (95% CI: 31.76 to 36.01 °C), respectively. At 72 h postinduction, the maximum temperature of the CNTL goats’ hooves were significantly lower (P = 0.006) than both Amp-B and K-C goats, with temperatures of 25.99 °C (95% CI: 23.73 to 28.25 °C), 30.80 °C (95% CI: 28.67 to 32.93 °C), and 30.13 °C (95% CI: 28.01 to 32.26 °C), respectively.

Figure 2.

Mean (±SE) maximum temperatures (°C) from infrared thermography for goats with induced lameness using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6).

Table 3.

Overall mean (95% Confidence Interval) maximum infrared thermography temperature (°C) of the left hind foot and mean temperatures (°C) by time point for goats with induced lameness using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6)

	LS1 means (95% confidence interval)			P-values
Parameter	Amp-B	K-C	CNTL	Treatment	Time	Treatment × time
Maximum infrared thermography, °C	35.16 (33.57–36.7)	35.37 (33.78–36.95)	32.30 (30.97–34.14)	0.03	<0.001	0.11
Baseline	35.38a   (33.26–37.51)	35.58a   (33.46–37.71)	34.97a   (32.84–37.09)	0.92
4 h	34.65a   (32.52–36.78)	36.10a   (33.97–38.23)	32.93a   (30.81–35.06)	0.12
6 h	37.18a   (35.06–39.31)	37.37a   (35.24–39.43)	35.68a   (33.56–37.81)	0.47
12 h	37.65a   (35.52–39.78)	37.85a   (35.72–39.98)	36.62a   (34.49–38.74)	0.68
24 h	35.99a   (33.73–38.25)	36.67a   (34.41–38.93)	31.62b   (29.49–33.74)	0.003
48 h	34.43a   (32.31–36.56)	33.88a   (31.76–36.01)	30.07b   (27.94–32.19)	0.01
72 h	30.80a   (28.67–32.93)	30.13a   (28.01–32.26)	25.99b   (23.73–28.25)	0.006

¹Least square means.

^a,bDifferent superscript letters within row, means differ (P ≤ 0.05).

Pressure Mat Gait Analysis

Data for PMT are summarized in Table 4. Pressure mat analysis for left and right rear limbs are shown in Figure 3.

Table 4.

Mean (± SE) outcome measures from pressure mat gait analysis in goats with induced lameness using 1 mL of 10 mg/mL amphotericin B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6)

	LS1 means (± SE)			P-values
Parameter	Amp-B	K-C	CNTL	Treatment	Time	Treatment × time
Left hind foot
Stance time, s	0.25a   (0.03)	0.22a   (0.03)	0.29a   (0.03)	0.26	0.0006	0.37
Stride length, cm	84.63a   (7.99)	73.88a   (7.99)	99.30a   (7.99)	0.11	0.05	0.64
Force, kg	11.76a   (2.35)	9.61a   (2.35)	20.43b   (2.35)	0.013	0.001	0.07
Impulse, kg × s	2.24a   (0.34)	1.91a   (0.34)	3.63b   (0.34)	0.007	<0.0001	0.13
Contact pressure, kg/cm2	3.85a   (0.42)	3.52a   (0.42)	4.98a   (0.42)	0.06	0.003	0.21
Contact area, cm2	2.62a,b   (0.39)	2.11a   (0.39)	4.03b   (0.39)	0.009	<0.0001	0.0003
Right hind foot
Stance time, s	0.39a   (0.05)	0.41a   (0.05)	0.31a   (0.05)	0.40	0.39	0.97
Stride length, cm	86.90a   (5.57)	80.72a   (5.57)	94.99a   (5.57)	0.22	0.48	0.37
Force, kg	17.52a   (1.89)	17.63a   (1.89)	20.29a   (1.89)	0.52	0.006	0.45
Impulse, kg × s	4.56a   (0.65)	5.17a   (0.65)	3.86a   (0.65)	0.39	0.02	0.91
Contact pressure, kg/cm2	4.87a   (0.27)	5.09a   (0.27)	5.11a   (0.27)	0.78	0.70	0.54
Contact area, cm2	3.61a   (0.28)	3.50a   (0.28)	3.88a   (0.28)	0.61	0.009	0.92

¹Least square means.

^a,bDifferent superscript letters within row, means differ (P ≤ 0.05).

Figure 3.

Mean (± SE) for the following: (A) left hind force (kg); (B) right hind force (kg); (C) left hind impulse (kg × s); (D) right hind impulse (kg * s); (E) left hind contact pressure (kg/cm²); (F) right hind contact pressure (kg/cm²); (G) left hind contact area (cm²); (H) right hind contact area (cm²) for goats with induced lameness using 1 mL of 10 mg/mL amphotericin B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6).

Stance time. There were no treatment group differences in stance time for left hind (P = 0.26) and right hind (P = 0.40) limbs. Mean stance times for left hind limbs of Amp-B, K-C, and CNTL groups were 0.25 s (95% CI: 0.18 to 0.31 s), 0.22 s (95% CI: 0.15 to 0.28 s) and 0.29 s (95% CI: 0.23 to 0.36 s), respectively. Mean stance times for right hind limbs were 0.39 s (95% CI: 0.28 to 0.51 s), 0.41 s (95% CI: 0.29 to 0.52 s), and 0.31 s (95% CI: 0.19 to 0.42 s) for Amp-B, K-C, and CNTL groups, respectively.

Stride length. Stride length is the distance measured between the posterior heel of two consecutive foot falls. There were no differences in stride length for left hind (P = 0.11) and right hind (P = 0.22) limbs between treatment groups. Mean stride lengths for left hind limbs were 84.63 cm (95% CI: 67.59 to 101.67 cm), 73.88 cm (95% CI: 56.84 to 90.92) and 99.30 cm (95% CI: 82.26 to 116.34 cm) for Amp-B, K-C, and CNTL groups, respectively. Mean stride lengths for right hind limbs of Amp-B, K-C, and CNTL groups were 86.90 cm (95% CI: 75.03 to 98.76 cm), 80.72 cm (95% CI: 68.85 to 92.58 cm), and 94.99 cm (95% CI: 83.12 to 106.86 cm), respectively.

Force. Force is the maximum force applied for each step. There was evidence of treatment effects in the amount of force on the left hind limb between Amp-B [11.76 kg (95% CI: 6.75 to 16.76 kg)], K-C [9.61 kg (95% CI: 4.60 to 14.61 kg)], and CNTL [20.43 kg (95% CI: 15.43 to 25.44)] groups (P = 0.013). The Amp-B and K-C goats applied lower force than CNTL goats at 6, 12, and 24 h posttreatment induction. At 6 h postinduction CNTL goats had a mean force of 24.57 kg (95% CI: 18.52 to 30.62 kg) compared to 10.69 kg (95% CI: 4.63 to 16.76 kg) and 5.46 kg (95% CI: −0.60 to 11.51 kg) for Amp-B and K-C goats, respectively (P < 0.001). Mean contact force was 20.21 kg (95% CI: 14.15 to 26.26 kg), 9.47 kg (95% CI: 3.41 to 15.52 kg), and 6.43 kg (95% CI: 0.37 to 12.48 kg; P = 0.006) at 12 h postinduction for CNTL, Amp-B, K-C goats, respectively. At 24 h postinduction, Amp-B [8.77 kg (95% CI: 2.71 to 14.82 kg)] and K-C [6.09 kg (95% CI: 0.04 to 12.14)] goats had lower (P = 0.005) measured force than CNTL goats [20.03 kg (95% CI: 13.98 to 26.09 kg)].

There were no differences between treatment group for contact force in the nonlame right rear foot (P = 0.22). The mean force applied was 86.90 kg (95% CI: 75.03 to 98.76 kg), 80.72 kg (95% CI: 68.85 to 92.58 kg), and 94.99 kg (95% CI: 83.12 to 106.86 kg) for Amp-B, K-C, and CNTL goats, respectively.

Impulse. Impulse is the maximum force applied per unit of time. There were treatment effects on the impulse of the left hind limb. The overall mean impulse for Amp-B, K-C, and CNTL groups were 2.24 kg × s (95% CI: 1.51 to 2.98 kg × s), 1.91 kg × s (95% CI: 1.18 to 2.64 kg × s) and 3.63 kg × s (95% CI: 2.90 to 4.37 kg × s) (P = 0.007). The mean impulse of Amp-B [1.32 kg × s (95% CI: 0.10 to 2.55 kg × s)] and K-C [0.93 kg × s (95% CI: −0.29 to 2.16 kg × s)] groups were lowest at 12 h postinduction. These were lower than the CNTL [3.64 kg × s (95% CI: 2.41 to 4.87 kg × s)] at 12 h (P = 0.005).

There was no evidence of treatment effect on impulse of the contralateral nonlame right hind limb. The overall mean impulse for Amp-B, K-C, and CNTL groups were 4.56 kg × s (95% CI: 3.17 to 5.95 kg × s), 5.17 kg × s (95% CI: 3.78 to 6.56 kg × s), and 3.86 kg × s (95% CI: 2.47 to 5.25 kg × s) (P = 0.39). There was evidence of a time effect (P = 0.02) with the mean impulse for all groups peaking at 6 h postinduction [5.07 kg × s (95% CI: 4.04 to 6.11 kg × s)] and low at 48 h [3.50 kg × s (95% CI: 2.47 to 4.53 kg × s)].

Contact pressure. Contact pressure is the peak pressure applied for each foot fall on the mat. Overall mean contact pressure for the left rear limbs was lower in Amp-B [3.85 kg/cm² (95% CI: 2.96 to 4.74 kg/cm²)] and K-C [3.52 kg/cm² (95% CI: 2.63 to 4.41 kg/cm²)] goats compared to CNTL 4.98 kg/cm² (95% CI: 4.09 to 5.87 kg/cm²)] goats (P = 0.007). There was evidence of a time effect with peak contact pressure recorded at baseline [4.89 kg/cm² (95% CI: 4.19 to 5.61 kg/cm²)] and lowest pressure at 12 h postinduction [3.37 kg/cm² (95% CI: 2.66 to 4.08 kg/cm²; P < 0.003)].

There were no differences between treatment groups for contact area in the nonlame right rear foot (P = 0.78). The overall mean contact area was 4.87 kg/cm² (95% CI: 4.30 to 5.44 kg/cm²), 5.09 kg/cm² (95% CI: 4.52 to 5.66 kg/cm²) and 5.12 kg/cm² (95% CI: 4.55 to 5.69 kg/cm²) for Amp-B, K-C, and CNTL goats, respectively.

Contact area. Contact area is the total area in contact with the pressure mat for each foot fall. The overall contact area of the left hind hooves of goats in Amp-B [2.62 cm² (95% CI: 1.90 to 3.45 cm²)] and K-C [2.11 cm² (95% CI: 1.29 to 2.94 cm²)] groups were lower than CNTL goats [4.03 cm² (95% CI: 3.21 to 4.86 cm²); P = 0.009]. Additionally, there was evidence of a time effect (P < 0.001) and treatment × time interactions (P = 0.0003). Goats in Amp-B and K-C groups had lower mean contact areas than those taken before lameness induction. The mean contact area prior to lameness induction was 3.64 cm² (95% CI: 2.66 to 4.62 cm²) for Amp-B goats and 3.70 cm² (95% CI: 2.72 to 4.68 cm²) for K-C goats. Both Amp-B and K-C groups had contact pressure nadirs at 24 h postinduction with mean contact areas of 1.73 cm² (95% CI: 0.76 to 2.71 cm²) and 1.44 cm² (95% CI: 0.46 to 2.42 cm²), respectively.

Evidence of significant treatment differences were observed at 6, 12, 24, and 48 h postinduction with CNTL goats having higher contact pressure. At 6 h postinduction CNTL goats had mean contact pressure of 4.36 cm² (95% CI: 3.39 to 5.34 cm²) compared to Amp-B and K-C goats of 2.57 cm² (95% CI: 1.59 to 3.54 cm²) and 1.14 cm² (95% CI: 0.16 to 2.12 cm²; P = 0.0002). Mean contact pressures for the Amp-B, K-C, and CNTL groups at 12 h were 1.84 cm² (95% CI: 0.87 to 2.82 cm²), 1.46 cm² (95% CI: 0.48 to 2.43 cm²) and 4.12 cm² (95% CI: 3.14 to 5.10 cm²; P = 0.0008). Mean contact pressures at 24 h were 1.73 cm² (95% CI: 0.76 to 2.71 cm²), 1.44 cm² (95% CI: 0.46 to 2.42 cm²), and 4.31 cm² (95% CI: 3.33 to 5.29 cm²; P = 0.0002) for the Amp-B, K-C, and CNTL groups, respectively. At 48 h postinduction, the mean contact pressure was 2.69 cm² (95% CI: 1.17 to 3.66 cm²), 2.29 cm² (95% CI: 1.31 to 3.27 cm²), and 4.10 cm² (95% CI: 3.12 to 5.08 cm²) for the Amp-B, K-C, and CNTL groups, respectively (P = 0.03).

Plasma Cortisol

Plasma cortisol concentrations are shown in Figure 4 and summarized in Table 5. There was no evidence of differences in mean CORT concentrations between treatment groups. The mean CORT concentrations were 17.67 ng/mL (95% CI: 10.02 to 20.31 ng/mL), 16.21 (95% CI: 8.57 to 23.86 ng/mL), and 10.25 ng/mL (95% CI: 2.61 to 17.89 ng/mL; P = 0.44) for Amp-B, K-C, and CNTL goats, respectively. There was evidence of a time effect (P = 0.0009) and a treatment × time interaction (P = 0.003). At 4 h postinduction, Amp-B [14.62 ng/mL (95% CI: 3.08 to 26.16 ng/mL)] and K-C groups [21.18 ng/mL (95% CI: 9.64 to 32.73 ng/mL)] had higher (P = 0.005) mean CORT concentrations compared to CNTL group [2.35 ng/ml (95% CI: −9.20 to 13.89 ng/mL)]. Mean CORT concentrations remained higher (P = 0.01) in Amp-B [30.28 ng/mL (95% CI: 18.74 to 41.83 ng/mL)] and K-C groups [40.13 ng/mL (95% CI: 28.59 to 51.68 ng/mL)] compared to CNTL group [8.27 ng/ml (95% CI: −3.27 to 19.82 ng/mL)] at 6 h postinduction. Evidence of treatment group differences were no longer seen (P = 0.61) at 12 h postinduction as mean CORT concentrations were 19.19 ng/mL (95% CI: 7.64 to 30.74 ng/mL) for the Amp-B goats, 20.30 ng/mL (95% CI: 8.75 to 31.84 ng/mL) for the K-C goats and 10.51 ng/mL (95% CI: −1.04 to 22.05 ng/mL) for CNTL goats.

Figure 4.

Mean (± SE) plasma cortisol concentrations (ng/mL) for goats with induced lameness using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6).

Table 5.

Overall mean (95% Confidence Interval) plasma cortisol concentrations (ng/mL) and mean cortisol concentrations (ng/mL) by time point for goats with induced lameness using 1 mL of 10 mg/mL amphotericin-B solution (Amp-B; n = 6), 1 mL of a 5% kaolin and 5% carrageenan mixture (K-C; n = 6), or sham induction using 1 mL of physiological saline (CNTL; n = 6).

	LS1 means (95% confidence interval)			P-values
Parameter	Amp-B	K-C	CNTL	Treatment	Time	Treatment × time
Cortisol, ng/mL	17.67 (10.02–25.31)	16.21 (8.57–23.86)	10.25 (2.61–17.89)	0.44	0.0009	0.003
Baseline	16.29a   (4.75–27.84)	7.02a   (–4.52–18.57)	18.87a   (7.32–30.41)	0.50
4 h	14.62a   (3.08–26.16)	21.18a   (9.64–32.73)	2.35b   (–9.20–13.89)	0.005
6 h	30.28a   (18.74–41.83)	40.13a   (28.59–51.68)	8.27b   (–3.27–19.82)	0.01
12 h	19.19a   (7.65–30.74)	20.30a   (8.75–31.84)	10.51a   (–1.04–22.05)	0.61
24 h	20.50a   (8.96–32.05)	13.73a   (2.19–25.28)	10.91a   (–0.63–22.46)	0.66
48 h	12.68a   (1.13–24.22)	6.15a   (–5.40–17.69)	6.88a   (18.42)	0.34
72 h	10.09a   (–1.45–21.63)	4.98a   (–6.56–16.53)	13.97a   (2.42–25.51)	0.23

¹Least square means.

^a,bDifferent superscript letters within row, means differ (P ≤ 0.05).

DISCUSSION

Lameness is an important animal welfare concern in ruminants. Understanding animal response to pain is critical for the development of appropriate treatment modalities. The objectives of this pilot study were to evaluate two different lameness models and measures for determining individual response to induced lameness. This study demonstrated the successful induction of acute, temporary lameness in goats after intraarticular injection of Amp-B or K-C. Lameness was subjectively and objectively characterized using VLS, IRT, PMT, and CORT. Behavioral changes were characterized between lame and control animals.

Amphotericin-B and K-C induced lameness models have been studied in other species (McIlwraith et al., 1979; Schulz et al., 2011; Coetzee et al., 2014; Muley et al., 2016). In the present study, Amp-B and K-C produced a controlled, sustained and transient lameness in most goats (10/12 goats). The severity of lameness peaked at 4 h postinduction. Resolution of lameness was short for most animals but variable between treatment groups. Four of the lame animals were sound by 72 h (Amp-B: n = 2, K-C: n = 2), five were sound by 96 h (Amp-B: n = 3, K-C: n = 2) and one with protracted lameness was not sound for 10 d (K-C: n = 1). Such findings are consistent with onset, severity and duration of lameness seen in cattle when Amp-B has been used to induced lameness (Schulz et al., 2011; Coetzee et al., 2014). The onset and duration of lameness seen in Amp-B goats and cattle is much shorter compared to horses (McIlwraith et al., 1979) supporting the theory that there are species specific differences in response to these lameness protocols. However, in the present study, 6 of 12 animals (Amp-B: n = 3; K-C: n = 3) met criteria for rescue therapy, which likely truncated duration and severity of induced lameness. Six of the twelve goats received a single intravenous injection of flunixin meglumine (2.2 mg/kg) at 24 h postlameness induction. One of the six goats (in the K-C treatment group), continued to display severe lameness requiring two additional doses of flunixin meglumine. Two goats (Amp-B: n = 1, K-C: n = 1) failed to develop lameness. The authors suspect that inadvertent periarticular rather than intraarticular injection in these two goats resulted in lack of lameness development. Additional studies are needed to better characterize the duration of lameness induced with either Amp-B or K-C.

Intraarticular injection with K-C has been shown to produce reliable acute arthritis in primates, cats and rats (Muley et al., 2016). Kaolin causes physical irritation to the joint while carrageenan stimulates activation of TLR 4 and subsequent leukocyte infiltration and inflammation (Muley et al., 2016). Due to the abrasive nature of K-C, the authors of this study predicted the severity and duration of lameness in goats receiving K-C would be greater than Amp-B. While there were no statistically significant differences between Amp-B and K-C treated animals, there were notable clinical differences. In this study, K-C compared to Amp-B treated goats demonstrated more severe lameness characterized by higher VLS, increased behavioral responses (tail wagging), and decreased placement (lower force, pulse, and contact area) of the affected limb on the pressure mat. The lack of statistical significance is likely due to the small number of animals in this study. Additionally, K-C was a highly viscous substance that was difficult to inject intraarticularly compared to Amp-B. Due to the severity of lameness induced by K-C and difficulty in intraarticular injection, the authors of this study suggest Amp-B may be a more reliable lameness model.

Lameness causes measurable changes in ruminant behavior, including more time spent lying, a decrease in feeding behavior and altered social interactions (Galindo and Broom, 2002; Bach et al., 2007; Ito et al., 2010; Blackie et al., 2011). Goats in pain may also appear restless, vocalize, and be more attentive to the area causing pain (Ajadi et al., 2012). A detailed ethogram describing behavioral changes associated with lameness in goats has not been developed. For this study, the authors used a modification of a previously described ethogram for goat behavior (Walker et al., 2015). Animals in both Amp-B and K-C groups demonstrated several consistent behavioral changes associated with lameness with one notable exception: animals in the K-C group displayed significantly more tail wagging behavior compared to Amp-B and CNTL animals. An increase in tail wagging after a painful event, such as castration or dehorning, has been observed in other livestock species, including piglets, lambs, and calves (Robertson et al., 1994; Graf et al., 1999; Hay et al., 2003; Rault et al., 2014; Jongman et al., 2016; Viscardi and Turner, 2018). This behavioral response suggests the induction of lameness with K-C may have been a more painful method for goats. However, this was a small group of goats and further studies using larger numbers of goats are warranted to better characterize caprine behavioral responses to pain.

Objective measures of lameness evaluation in small ruminants are lacking. Lameness is most commonly graded using a VLS. Such scoring systems are subjective resulting in inter-observer differences (Keegan et al., 2010). In the present study, a single veterinarian, with extensive clinical experience in lameness identification of food animals, blinded to treatment, performed all VLS. A single observer was used to eliminate any inter observer variability that may have occurred with multiple people scoring the goats. However, relying on a single examiner does not rule out intra observer variability which cannot be eliminated in this study. The VLS in this study was used to determine obvious differences in lameness between treatment groups.

Objective measures of lameness detection used in this study included IRT and PMT. Infrared thermography was recently validated for detecting lameness associated with foot lesions in sheep (Byrne et al., 2019). Obvious temperature differences were noted between diseased and healthy hooves, indicating IRT could be reliably used to detect lameness in sheep (Byrne et al., 2019). Our study identified similar thermal changes between diseased (Amp-B and K-C) hooves compared to CNTL suggesting IRT was a reliable method for detecting induced lameness in goats.

Pressure mat gait analysis was recently evaluated as a tool for characterizing gait in healthy goats (Rifkin et al., 2019). Authors of that study found that the PMT objectively characterized the gait of goats but there was some difficulty in collecting a complete data set from each goat. In the present study, goats were easily trained to walk across the PMT and there was no difficulty collecting a complete data set for each goat. The PMT system used in this study had a larger recording area and two steps per foot were readily obtained. The PMT used in this study had a custom-built alley that helped to guide the goats through the PMT which likely facilitated data collection.

Observed differences in pressure mat outcomes between the CNTL goats and those in the Amp-B and K-C groups indicates PMT is an objective tool for lameness evaluation. Differences in the lame limb (left hind) were observed, but not for the sound right hind limb. Additionally, time interactions were observed where decreases in contact force, area and pressure were seen. Furthermore, goats requiring rescue analgesia at 24 h had lower contact force and area compared to their counterparts not needing analgesia.

Plasma cortisol has been used as a measure of stress and inflammation associated with lameness induction in other livestock species (Coetzee et al., 2014). In this study, goats in the two lameness groups (Amp-B and K-C) had elevated CORT levels at 4 and 6 h postlameness induction. This is consistent with the changes in VLS and PMT data. There is little data available to describe CORT concentrations in goats with lameness. In cattle, lameness has been associated with increases in CORT concentrations (Coetzee et al., 2014; Kleinhenz et al., 2019).

To the authors knowledge, this is the first pilot study describing a lameness model for use in meat goats. Objective measures of lameness evaluation are lacking in goats. Using a modified behavioral scoring system, IRT, CORT, and PMT this study was able to provide a unique behavioral assessment and objective measures of individual goat response to pain associated with lameness. This pilot study provides much needed information regarding lameness in goats required for future investigations.

Despite successful lameness induction of goats in this study, the described protocol warrants further investigation. As mentioned throughout, the small number of animals likely contributed to limited statistical significance seen in behavior and objective lameness measures. The study should be repeated with a larger group of individuals. There are demonstrated differences in response to lameness between species (Schulz et al., 2011; Coetzee et al., 2014, McIlwraith et al., 1979). In the present study, a single breed of similarly aged goats was used. Specifically, future investigations should include a variety of breeds (meat vs. dairy), sexes, and ages. Duration of lameness extended beyond the 72 h time point predetermined for monitoring in the present study. Future studies should continue lameness monitoring for at least 96 h. A final consideration for refinement of the current lameness model should be modification of the dose of Amp-B and K-C. For example, the K-C was very viscous and the small 1 mL volume used was difficult to inject into the very small distal interphalangeal joint. Due to the small size and viscosity, titration of the dose based off animal size may be warranted.

The objectives of the current study were to asses two different lameness models and measures for quantifying individual response to induced lameness. Amphotericin-B and K-C were able to induce an acute, transient lameness in meat goats. Lameness was readily characterized using VLS, IRT, and PMT. The present study is the first pilot study evaluating IRT together with PMT in meat goats with induced lameness. The small number of animals in each treatment group and the large number of animals requiring treatment were limitations in this study. Future studies are needed to better characterize the lameness induced by Amp-B and K-C.