Behavioural and physiological measures of welfare of pregnant mares fitted with a novel urine collection device

https://doi.org/10.1016/j.applanim.2004.12.005Get rights and content

Abstract

An alternative collection procedure for pregnant mare urine (PMU) has been developed in Australia, which allows mares to be loose housed either in indoor stables or outdoors in paddocks, rather than in tether stalls as is common practice in Canada and North Dakota. The present study examined the welfare risks to mares of collecting urine using this alternative procedure. The study involved 24 pregnant mares at 3–5 months of gestation. The mares were allocated to two treatments: 12 mares were fitted with the pregnant mare urine collection device and 12 mares acted as controls. All mares were housed in two large paddocks during the day, in two groups of 12. During the night, all mares were housed in six small enclosures in groups of four, while the PMU treatment mares wore the collection device.

Mares wearing the PMU collection device showed little or no behavioural change relative to the control mares. While brief bouts of forward lifting of the hind leg in the region of the PMU collection device were observed in some mares on day 1, this disappeared by day 2. Observations on the time budgets of behaviour in weeks 3 and 6–7 indicated similar patterns of behaviour in the two treatments. The only difference in behaviour was in lying behaviour. PMU collection mares spent less (P < 0.01) time lying during the night, with fewer horses observed lying down in the PMU collection treatment (4/11 versus 10/12). It is unclear what implication this may have on horse welfare since there was no evidence of a rebound in this behaviour when the device was removed. Based on heart rate and cortisol concentrations measured on days 1 and 2 of study, there was no evidence of acute stress associated with fitting the device. Limited measurements on salivary cortisol concentrations at weeks 6–7 of the study provided no evidence of an increase in basal cortisol concentrations or an increase in the sensitivity of the adrenal cortex to ACTH in the PMU collection mares. Therefore, it is concluded that wearing the PMU collection device at night for five nights per week for a total of 6–7 weeks does not pose a serious challenge to the welfare of pregnant mares.

Introduction

There is a worldwide market for pregnant mare urine (PMU) to produce oestrogenic products used in human hormone replacement therapy. The segment of the horse industry presently involved in collecting PMU is concentrated in Canada and the USA (North Dakota), where more than 35,000 mares are housed on about 450 farms. Horses are housed in tether stalls for 5–6 months of the year, during which time some horses may have very limited opportunities for exercise outside the stall. In these PMU collection situations, horses wear a loose fitting, light-weight pouch, held in position by a system of pulleys. As a consequence of the restriction of movement of the mares, animal protection groups have voiced concern about the welfare implications of the housing system for this PMU collection procedure. An alternative collection procedure has been developed in Australia by Thorgard Pty. Ltd., which allows mares to be loose housed either in indoor stables or outdoors in paddocks. The present experiment examined the welfare risks to mares of collecting urine using this alternative procedure. The control for this examination was mares in which the collection device was not fitted but which were housed and handled in similar loose conditions. The welfare implications of the treatments were examined by measuring the behaviour and stress physiology of the horses in both the short and long term. The company intends to collect urine only during the evening and night and to limit collection to 5 days a week out of welfare considerations for the mares. This collection protocol was followed during the duration of the trial.

To study the short-term effects on the welfare of mares in the present study, cortisol concentrations, heart rate and behaviour before and after fitting the device were measured. After 6–7 weeks of treatment, basal plasma cortisol concentrations, adrenal responsiveness to ACTH and general behavioural activity were measured. Salivary cortisol is predictive of plasma cortisol in many species (Fell et al., 1985, Mendl et al., 1991, Mendl et al., 1992) and this sampling procedure, which is generally non-traumatic, was used in this study to measure both basal salivary cortisol concentrations and salivary cortisol concentrations in response to a challenge with ACTH. The responsiveness of the adrenal cortex to ACTH administration may indicate a state of chronic stress when the secretion of cortisol is increased in response to an injection with ACTH (Terlouw et al., 1997). A long-term inability to perform a behaviour that is highly motivated (such as lying down, grooming and exploration) may lead to a stereotypy at the time of the restriction or a rebound in behaviour when the restriction is removed (Mueller et al., 1999). Specific mild and repeated localised stimulation that is disturbing may lead to excessive rubbing and in turn injury, and thus, both behaviour and injuries were also monitored in the present study. Pressure and uneven weight distribution may result in localised pain of skin and muscle and even affect the spinal column, and therefore, a chiropractic examination was undertaken on the study horses.

Section snippets

Materials and methods

The study was conducted in March and April (autumn) at a horse stud in Victoria, Australia. A total of 24 mares of several breeds were used, although most were thoroughbreds and Australian Stock horses. They were sourced from three different breeders and ranged in age from 4 to 23 years old. All mares were weighed and confirmed pregnant, and were 3–5 months in gestation at the start of the study.

The mares were allocated to two treatments: 12 mares were fitted with the PMU collection device (PMU

Standard protocol

At 16:00 h, the horses were moved from the group paddocks to their night enclosures. Good quality meadow hay with a 60% ryegrass and 40% clover mix was placed in the night enclosures to facilitate movement of horses from the day paddocks. After all horses had entered the night enclosures, they were taken in pairs consisting of one control mare and one PMU collection mare by two people into an indoor fitting area in the same order every day. They were either fitted with a device (PMU collection

Veterinary inspection

One week prior to the start of the study, all horses underwent a visual inspection to determine pre-existing injuries and scars and an internal examination to confirm pregnancy. The veterinary inspection was repeated in weeks 6–7 to determine injuries and scars that may have been caused by the PMU collection treatment. The inspection was conducted by an independent veterinarian.

Chiropractic examination

The chiropractic examination was conducted by a qualified veterinarian and equine chiropractor. The examination

Measurements taken immediately after imposing treatments

Measurements that were taken immediately after imposing the two treatments were analysed using a paired comparison analysis of variance, with a pair taken to be the two mares simultaneously observed by one observer. The majority of measurements were transformed, prior to analysis, to reduce the skewness of the data. Results are presented as backtransformed means. Angular transformations were used for most behaviour observations. A log10 transformation was applied prior to analysis of the heart

Results

Two mares aborted during the trial, and both mares were in the PMU collection treatment. The mares were able to abort unassisted while wearing the collection device. The first mare aborted with twins 14 days after commencing the study due to placental insufficiency to carry twins. The foetuses were examined post-mortem at the Victorian Institute of Animal Science, Attwood, Australia. There was no evidence of infectious causes and the cause was considered unrelated to the device. This mare was

Heart rate

During the first fitting on day 1 the average heart rate over 15 min was not significantly (P > 0.05) different between the two treatments. There were also no significant treatment effects on average heart rate from 20 to 60 min (P > 0.05, Table 1). Heart rate measured while mares were taken to the fitting area and during fitting with the collection device in weeks 3 and 6 is presented in Table 3. Heart rates were significantly lower (P < 0.05) in the PMU collection treatment than in the control

Discussion

The measurements taken after the first imposition of the treatment and during the second night indicate that the horses had no serious problems adapting to wearing the collection device. The horses in the study were generally familiar with wearing rugs and saddles, and so the PMU collection equipment was probably not a highly novel experience. The significant increase in forward lifting of the hind leg in some horses fitted with the PMU collection device indicates some discomfort with this

Acknowledgements

Technical assistance of staff and students of the Animal Welfare Centre is acknowledged. In particular, the support of Adele Arnold, Elaine Leeson, Bruce Schirmer, Amanda Kobelt, Naomi Botheras and Michelle Edge is recognised. The salivary cortisol assays were conducted under the supervision of Dr. Ian McCauley and the statistical advice of Mr. K.L. Butler and the veterinary advice of Dr. Des James are acknowledged. The collection technology which is the subject of this welfare trial is owned

References (23)

  • Cited by (11)

    • Effects of grazing muzzles on behavior, voluntary exercise, and physiological stress of miniature horses housed in a herd

      2020, Applied Animal Behaviour Science
      Citation Excerpt :

      It is possible that allowing horses physical contact in group housing may lower SC. One assessment that could have enhanced interpretation of whether chronic stress was experienced by the study population is the application of an ACTH challenge test which has been used to observe chronic stress in mares fitted with urinary catchment devices (Jongman et al., 2005). In conclusion, this study demonstrated that muzzling miniature horses for either 10 or 24 h altered time spent performing behaviors but did not appear to induce physiological stress.

    • Salivary cortisol measurement in horses: immunoassay or LC-MS/MS?

      2020, Domestic Animal Endocrinology
      Citation Excerpt :

      In equine stress research, saliva sampling offers a simple and non-invasive method to assess the HPA axis in horses that are used to be handled by people. Salivary cortisol measurements have been used to investigate stress in horses during isolation [2], novel manipulations [3], weaning [4], transportation [5–8], hot iron branding and microchip implantation [9], training and competition [10–14], or during work in the presence of an audience [15]. Salivary cortisol has also been measured after HPA axis stimulation by means of an ACTH stimulation test.

    • Equine stereotypic behaviors: Causation, occurrence, and prevention

      2013, Journal of Veterinary Behavior: Clinical Applications and Research
      Citation Excerpt :

      Pregnant mare urine (PMU) industry farms collect the PMU to make estrogen supplements. The welfare of pregnant mares in the PMU could be an area of concern because of restrictive housing system in which horses have limited opportunity for movement and exercise (Jongman et al., 2005). However, lower incidences of stereotypies in tie-stall situations (e.g., PMU farms) have been reported compared with stable housing systems.

    • Agricultural Animal Welfare

      2013, Laboratory Animal Welfare
    • Pain and stress in crustaceans?

      2009, Applied Animal Behaviour Science
    View all citing articles on Scopus
    View full text