Bone and joint injuries in Thoroughbred racehorses typically originate in areas of intense loading and are attributable to fatigue damage accumulated under repetitive cyclical loading. This thesis will focus on an epidemiological understanding of the bone fatigue process as it relates to Thoroughbred training and racing. Factors investigated include workload, stride parameters, speed and distance worked. Specifically, this included an investigation of racehorse management, training and rest compared to previously published risk factors and how they relate to performance, stride characteristics of galloping horses during racing and how varying strides and race factors influence the estimated accumulation of bone damage over race starts. It was hypothesised that (i) there would be a large variation between trainers in the training and rest practices; (ii) greater workloads in training would not be associated with superior race performance; (iii) speed would be correlated with stride parameters, increasing proportionally more with stride duration than the number of strides (and therefore stride length) at higher speeds; (iv) stride parameters would differ by horse-and race-level factors; and (v) a higher percentage of bone fatigue life would be used for longer distance races, firmer track surfaces and higher classed races. These hypotheses were investigated by conducting a semi-structured in-person interview of 66 registered Thoroughbred trainers in Victoria, Australia; by analysing GPS and accelerometer data including speed and stride data from 25,245 Thoroughbred race starts in Tasmania, Australia; and by generating a mathematical model of bone fatigue accumulation using information from the stride parameter data. Australian Thoroughbred training programs include high volumes of galloping with more than half exceeding previously reported risk levels for fracture. Older age and longer intended race distances were associated with higher galloping workloads, and there was a large degree of variation between trainers in workloads, rest practices and surfaces used in training. Trainers who utilised the lowest and highest galloping workloads (at speeds of >13.3 or 13.3-14.3 m/s) as horses prepared for racing had fewer wins and places. There was substantial inter-horse variation in stride parameters, with speed predicting half or less of this variation. Male sex, greater race distance, better finishing position, and firmer track surfaces were associated with fewer strides per 200 m and longer stride durations. Females, older age, longer race distances, firmer track surfaces, and greater weights carried for high-classed races were associated with greater percentage fatigue life accumulated over each race start. There was a wide variety of strategies implemented by trainers in preparing and maintaining horses for racing and managing them between racing campaigns, yet this variation did not translate to how well horses performed. Variation was identified for stride parameters of horses galloping in races under different conditions, but further variation existed based on innate qualities of the individual horse, therefore future studies assessing risk for individual horses would benefit from the inclusion of stride characteristics. The future goal is to establish appropriate racehorse training and racing programs which allow for appropriate bone adaptation to withstand load whilst avoiding fatigue failure.

The tarsus or hock is a complex joint of the hind limb consisting of seven different bones and having various articulations. The tarsal bones are held together in intimate contact through the attachments of soft tissue such as ligaments, tendons and fascia. The tarsal joint is commonly involved in equine hind limb lameness and other pathological problems. Despite suggestions that tarsal conformation is a contributory factor to performance, lameness and orthopedic health, limited studies are available in the literature regarding tarsal bone morphometry, objective radiographic conformation or alignment changes in response to loading of the tarsus in horses, and the role of soft tissue in maintaining the joint stability and range of motion under loading is not known possibly due to insufficient measurable tarsal parameters. In addition to this, faulty conformation and uneven loading of the tarsal bones is believed to result in tarsal pathology, ligamentous ruptures and joint instability of the equine hind limb. The aims of this thesis were 1. To investigate the morphometric and radiographic anatomy of the equine tarsus and develop reliable measurable parameters that can be used to consistently and objectively measure tarsal conformation in horses. 2. To provide baseline data for the developed parameters from tarsal radiographs including the effects of rotations, and inter-rater repeatability. 3. To use developed parameters to assess the change of alignment of the tarsal joint under loading. 4. To investigate the potential role of the tarsal ligaments in the load redistribution within the tarsus, joint stability and range of motion of the tarsal joint. 5. To deduce the potential role of fascia in load redistribution through mechanical connectivity, internal composition (histological) and thickness variation proximal and distal to the tarsus. A pilot study on 10 cadaveric equine hind limbs from 7 adult horses, was used to specify the characteristics and identify reliable landmarks of the bones of the equine tarsal joint to be used for further investigation through radiology to assess the tarsal conformation. “Zero degree Lateromedial” (ZLM) and “Zero degree Dorsoplantar” (ZDP) views were established and 88 measurable parameters with validated anatomical landmarks were developed, including 39 parameters from the ZLM view and 49 parameters from the ZDP views, consisting of individual, as well as correlated angular and ratios parameters to avoid any effects of different tarsal size or magnification effects associated with the radiography technique. LM and DP views were used as the minimum required to gain a 3D appreciation of tarsal bone movements, while a complete investigation would require the inclusion of oblique views as well. Then subsequent studies established: i. baseline data of these parameters from tarsal radiographs. ii. effects of rotations, vertical and horizontal rotations of projection angles of the primary X-ray beam (radiography machine) on the developed tarsal measurements to established acceptable range of rotations. iii. assessment of the inter-rater reliability of the selected parameters. iv. changes in alignment of the tarsal bones under ex vivo loading. v.) Role of soft tissue like ligaments and fascia in loading and joint range of motion (during flexion, extension and incremental load. Increased load on the tarsal joint produced measurable radiographic changes in the alignment and direction of movement of the tarsal bones ex vivo. Six reliable parameters showed the potential to detect changes. Soft tissues such as ligaments were shown to play a significant role in maintaining joint stability and normal range of joint motion. Severing all major tarsal ligaments caused significant changes in tarsal flexion (P < 0.001). Cutting the plantar aspect of the long plantar ligament caused the largest significant change in tarsal extension. Both transverse and longitudinal arrangements of intertarsal ligaments within the tarsus and the arrangement of major tarsal ligaments external to the bones suggest a normal tightly packed alignment of the tarsus allows slight transverse movements of the tarsal bones during loading and prevents over extension and flexion of the tarsal joint, suggesting the existence of more complex mechanisms involving other soft tissue such as fascia that help in maintaining the loading stability and influence the joint range of motion. In consideration of this, this research was then directed towards the broader field of fascial anatomy. Fifteen hind limbs were obtained from adult horses of mixed breeds and an anatomical exploration of fascia by dissection was completed with the purpose being to lay the groundwork for understanding how tension is distributed from the crural region to the distal hindlimb. Characteristics such as relative fascial thickness, attachment points, connectivity and collagen fiber’s orientation, internal composition by histological examination were recorded and described to deduce their potential functional significance. In conclusion, the tarsal bone geomorphometric and loading data, along with information generated on the ligaments and fascia in this study, will allow reliable quantitative assessment of tarsal conformation and eliminate judgmental errors or variation between observers using subjective visual assessment for the tarsus. Locomotory control and the postural stability of the hind limb is largely dependent on the functional integration of different structures which is enabled via the fascial architecture providing mechanical connectivity. Overall, this data will improve our understanding of tarsal loading biomechanics and help in the understanding of the pathogenesis of injury and locomotor dysfunctions. The measurement protocols will require further investigations on large groups of different horse breeds for wider potential usefulness, adaptability and validation. The use of the developed parameters to differentiate between normal conformation and any faulty or pathological conformation will potentially allow investigators to identify possible relationships between morphological measurements and the occurrence of pathology.

Bone and joint injuries in Thoroughbred racehorses typically originate in areas of intense loading. Despite the increasing understanding regarding the pathogenesis of bone fatigue, early detection prior to fatigue fracture is difficult. This thesis focuses on understanding equine behaviour and its potential to reflect orthopaedic pain to assist in the early identification of horses at risk of injury. Previous research suggests an association between facial expression and the presence of equine lameness. In cattle, motion sensors have been validated to record recumbency behaviour used to detect disease without the need for continuous observation. This thesis aimed to investigate monitoring of behavioural patterns as predictors of injury in racehorses. In the first study, Thoroughbred racehorses were monitored at rest with motion sensors placed on forelimbs. Sensor data was collected, and an algorithm developed to identify certain behaviours. Algorithms were validated through video observation and calculation of algorithm sensitivity and specificity. There was excellent agreement for standing, lying, and for stepping behaviour patterns suggesting potential for long term, objective, remote monitoring of horses in the stable environments. In the second study, previously validated facial grimace scales were used to score pain from photographs of Thoroughbreds, captured from video recordings during trot up examinations. This method demonstrated horses with moderate mouth strain (HGS) and tense and extended upper lip (FEReq) were associated with horses less likely to be lame. Exposed sclera was associated with lameness in hand, at the trot. Cohen's kappa was used to measure the proportion of agreement between raters, with behaviours scored to more than two categories the kappa was ordinal weighted. Mean inter observer agreement was moderate, and the sum of facial scores showed no difference between lame and sound horses. There were few associations between lameness status and facial predictors, limiting the potential use of facial expression for prediction of orthopaedic pain during pre-race lameness examinations. In conclusion, motion sensors are sensitive to detect horse movements and have potential use in longitudinal monitoring of horse behaviour. Future research in this area includes prediction of behavioural indicators of musculoskeletal pain in racehorses that could contribute to the development of a risk-based injury prevention model. The facial grimace scales had limited ability to differentiate lame from ‘sound’ horses in this cohort of racehorses. External factors associated with racehorse management, including environmental stress, may be one reason facial grimace scales are not effective for the recognition of subtle lameness in this cohort.

The use of laparoscopic surgery for routine procedures such as ovariohysterectomy has been well described for dogs and is common in humans. Rabbits have been previously used as models for human laparoscopic surgery and training models for paediatric surgery, however reported use of clinical laparoscopy in rabbits is rare. There are concerns for use of laparoscopic surgery in rabbits due to the effects of the insufflation on ventilation and the risk of increased morbidity from the insufflation contributing to gastrointestinal stasis, a common and life-threatening complication of any surgery in rabbits. This study is designed to quantify and characterise the changes in the postoperative morbidity between open, insufflated and isobaric laparoscopy in healthy adult rabbits. The hypotheses were that use of isobaric laparoscopy will decrease the morbidity of ovariohysterectomy procedures compared to open and insufflated ovariohysterectomy at the expense of increased surgical time. Various investigations were performed over the research project, including a technical viability cadaveric study, a study describing the effects of isobaric and insufflated pneumoperitoneum on ventilatory capability and abdominal dimensions, a study describing the clinical implementation of a Rabbit Grimace Pain score and Behavioural Pain Score in the detection of postoperative pain, and a clinical trial assessing the effects of both laparoscopy methods and comparing them with open laparotomy for ovariohysterectomy. The overall findings of the study support the implementation of isobaric laparoscopy in the rabbit, and the use of laparoscopy in general as a method of reducing postoperative morbidity compared with equivalent laparotomy approaches.

The southern brown bandicoot (Isoodon obesulus obesulus), a small, ground-dwelling marsupial, is listed as ‘endangered’ under the Australian Environment Protection and Biodiversity Conservation Act 1999. While many factors contributing to population decline are understood, in particular predation and loss of high-quality, connected habitat, there is a lack of knowledge about other threatening processes, including disease, which has contributed to declines in other small mammal species. Toxoplasma gondii, a protozoan intracellular parasite, excreted into the environment by cats, has been shown to cause clinical disease, including death, in many small and medium sized captive and free-ranging marsupials, including the eastern barred bandicoot (Perameles gunnii). Little is known of its effect on southern brown bandicoot populations. This study aimed to investigate the significance of T. gondii to the health of southern brown bandicoot populations on the northern hinterland of Western Port, Victoria and methods of predicting probability of infection with environmental T. gondii. A series of necropsies was performed on 33 southern brown bandicoots collected opportunistically over a five-year period. The causes of death were identified as motor vehicle trauma (22); predation (4); ejected pouch young (2); drowning (1); pyometra (1); possible toxicity and stomach bloating through lentil ingestion (1) and unknown (2). Real-time qPCR was performed on tissues from 30 cases, all of which were negative for the presence of T. gondii DNA. A range of helminths and ectoparasites were collected and identified, most of which had been previously been reported in this species. However, a metastrongyloid helminth species, found by histopathology in the lungs, is reported for the first time in a southern brown bandicoot. To assess environmental contamination with T. gondii, two sites with different cat densities were compared. Seroprevalence of antibodies to T. gondii (n=24) using the Modified Agglutination Technique was performed on trapped southern brown bandicoots. No evidence of infection with T. gondii was found at either site. Molecular (qPCR) methods were used to measure T. gondii oocyst presence in soil samples (n= 594) and prevalence of T. gondii in tissues of rabbits (n=118) and mice (n=267). All tests were negative across both sites except for the presence of T. gondii in one rabbit (prevalence 0.85%). These results suggest that rabbits and mice may have the potential to be reliable sentinel species and inform conservation management of the probability of infection with T. gondii in small marsupials. The results from this study are consistent with the concept that opportunistic necropsy is a valuable strategy for passive disease and cause of death surveillance in native wildlife. No evidence was found that T. gondii was impacting the health of southern brown bandicoots, however, further longitudinal health surveys are necessary to determine the true prevalence of disease and causes of mortality. Further studies are recommended to confirm the effectiveness of mice and rabbits as potential sentinel species in a range of ecosystems, including those environments which have higher levels of contamination with T. gondii.

Spinal cord injury (SCI) is a common cause of neurological deficits in dogs and a major cause of morbidity and decreased quality of life in people. In dogs, the most common causes of SCI are intervertebral disc diseases, particularly 'type I' intervertebral disc herniation knowns as intervertebral disc extrusion (IVDE), which occurs predominantly in chondrodystrophic breeds. The IVDE commonly causes contusion and compression to the spinal cord, which in most severe cases leads to complete loss of voluntary motor functions, bladder control and pain sensation distal to the level of the lesion. For decades, researchers have used animal models in their studies of SCI. However, limited progress has been achieved in translating any beneficial findings from experimental animal models to human SCI. Moreover, experimental studies in rodent models may not closely represent clinical SCI in human and canine patients. Naturally occurring compressive and contusive lesion of SCI associated with IVDE in dogs may be more analogous to human disease than induced experimental animal models. Canine naturally occurring SCI could simulate the injury in human patients including; mechanism of damage, pathophysiological events, healing processes, functional recovery, measures of evaluating outcomes, and effectiveness of the candidate therapies. Therefore, recruiting dogs of naturally occurring IVDE associated with SCI has become a common practice in clinical research trials. To date, no therapeutic method has been shown to treat SCI successfully in dogs with severe lesions. Therefore, the long-term outcome in these cases is chronic paraplegia with loss of sensation distal to the level of the lesion. Since this is true in both human and canine patients, novel therapies must be investigated, including promising neural stem cells (NSCs) therapy.   The principal objective of this thesis was to study canine SCI and develop a reliable and reproducible paradigm for evaluating the effect of candidate therapies, using human neural precursor cells (hNPCs) in dogs as an example. This was approached by; 1) evaluating different MRI measures to monitor the spinal cord lesion in putatively naturally occurring canine spinal cord trauma, pre- and post- surgical or non-surgical treatments of IVDE, 2) ex vivo examination of hNPCs’ impact on inflammatory factors in stimulated whole blood (WB) samples from healthy dogs, 3) evaluating methods to measure responses to transplant of hNPCs into the spinal cord of the paraplegic dog with grade 5 SCI, as a pilot clinical study, 4) exploring the fate of hNPCs at 18 months post-xenotransplantation into spinal cord tissue (SCT) of a dog with grade 5 SCI, and 5) investigating the possibility of isolate and culture of canine NSCs from SCT of the dog. The findings of this thesis showed that the pathological and morphometric alterations of the objective and quantifiable parameters of the MRI measures of the spinal cord and vertebral canal post-SCI and following treatments are important prognostic indicators. The interaction effect of time and surgery is an important factor in decreasing spinal cord compression and vertebral canal narrowing. The results of the ex vivo study suggested that hNPCs could modulate the inflammatory responses in stimulated WB by reducing the production of tumour necrosis factor-alpha. The in vivo clinical pilot study concluded that neither adverse effects nor immunological reactions were developed following transplantation of hNPCs into the spinal cord of the dog. Although recovery of the cutaneous trunci muscle reflex was identified, no improvements of the deep pain sensation or locomotion and urinary functions were shown at the end of the 6 months observation period post-therapy. The immunohistochemical findings revealed that the grafted hNPCs had survived and migrated, after xenotransplantation into SCT of the host dog. The gross and histopathological examination showed neither significant pathological changes nor tumour formation at 18 months following xenotransplantation in the dog with grade 5 SCI. This research work reported that it is possible to isolate, and culture canine derived NSCs from SCT. Therefore, this shows promise for future research using of dog-to-dog allotransplantation of canine derived NSCs, which may minimise the associated complications of xenotransplantation.   In summary, the observations of the studies in this thesis offer insight into the methods developed to evaluate potential benefits of using hNPCs for transplantation therapy in dogs with grade 5 SCI. The current research work helps understanding of the pathological and morphometrics alterations of spinal cord lesion pre and post treatment of IVDE. The findings are of value in understanding the mechanism of potential therapeutic effects of hNPCs, including the anti-inflammatory effect and the fate of the cells following transplantation into the spinal cord of the host. Finally, this research represents an initial step towards investigating the efficacy and potential benefits of hNPCs in a preclinical study in a large number of dogs with naturally occurring SCI.  

Infectious laryngotracheitis virus (ILTV) is an alphaherpesvirus that causes acute respiratory disease in chickens. This disease causes economic loss in poultry industries worldwide and is a major concern for animal health and welfare. Although, ILTV vaccines are in use to control disease, biosecurity breaches and continuous evolution of the ILTV genome make outbreak prevention difficult. Genomic recombination plays a role in diversification of herpesvirus genomes and has been suggested to be an important alternative evolutionary mechanism in herpesviruses such as ILTV. In previous studies of ILTV, two new genotypes of virulent field strains were shown to be independent recombinants derived from distinct attenuated commercial ILTV vaccines. These new strains became the dominant field viruses responsible for widespread and severe disease outbreaks in Australian poultry flocks. This project aimed to determine the frequency of these recombination events and the conditions that can facilitate recombination using in vitro, in ovo and in vivo studies. In these series of studies, a high-throughput and cost-effective method for detection of recombinant ILTVs was developed and was used to analyse the viruses generated under multiple infection conditions. The results from these analyses indicated that genomic recombination between vaccine strains of ILTV is a frequent event and takes place under a broad range of conditions. De novo mutation was detected in addition to genomic recombination. The capacity of ILTV vaccines to allow secondary infection by another ILTV vaccine (i.e. superinfection) was investigated at both cellular and natural host levels. The result of this investigation showed that superinfection can occur after administration of ILTV vaccines, providing the virus with ample opportunity for genomic recombination. The outcomes of these studies include advances in fundamental understanding about the factors that lead to ILTV recombination.

Methods of monitoring training load are widely practiced amongst today’s elite, sub-elite and amateur athletes. Monitoring training load can range from simple, cost effective methods to complex, expensive methods. The aim of this research was to investigate different methods of monitoring training loads in young Thoroughbred racehorses in pre-training. Simple, cost effective measurements of proximal hoof circumference and hoof toe angle were investigated as a tool to monitor how the hoof responses to the loads experienced in the first two pre-training preparations in young Thoroughbred racehorses. Left and right proximal hoof circumference (n = 42) and hoof toe angle (n = 39) measurements were obtained weekly. During the first pre-training preparation, a significant reduction was recorded in both proximal hoof circumference and hoof toe angle. In contrast, proximal hoof circumference did not change significantly during the second pre-training preparation, whereas hoof toe angle showed a smaller significant decrease. The advancement of modern technology has allowed for improvements in the simplicity of radiographic measurements. However, the collection of radiographic images is still costly to the Thoroughbred racehorse owner. The response of the third metacarpal bone shape and size to training loads can be easily evaluated through the frequent collection of radiographic images. Firstly, weekly left and right mediolateral and lateromedial (respectively) radiographs of the third metacarpal bone were collected and radiographic measurements of bone shape and size were analysed with proximal hoof circumference. Secondly, measurements of supracondylar modelling were obtained the distal end of these third metacarpal bone radiographs. Fifty-four Thoroughbred racehorses were assessed during the first and second pre-training preparations for both sets of measurements. Forty-four Thoroughbred racehorses were assessed at follow-up supracondylar modelling measurement taken from radiographs at 1.5 or 2.5 years from the original measurements. The results showed a moderately strong positive correlation between supracondylar modelling and the number of days in training. Heart rate recovery (HRR) is a simple and effective method of analysing changing fitness levels in Thoroughbred racehorses. Predominantly, research to observe HRR values has been conducted in the laboratory rather than in the field. In the current study, HRR was assessed in thirty-six Thoroughbred racehorses during the first and last sessions of one pre-training preparation. Heart rate data were collected at peak heart rate, 60- and 120-seconds post peak heart rate, at the end of the training session, and 60- and 120-seconds post training. Heart rate recovery was calculated by subtracting the heart rate value at 60-seconds post peak heart rate from the peak heart rate. A significant difference was found between HRR at the first and last HRR assessments. Heart rate recovery assessments in the field may be more difficult to obtain than HRR assessments in the laboratory due to the separation of the trainer and the Thoroughbred racehorse. Therefore, the use heart rate monitors and GPS devices to collect and store HRR values may be more beneficial as it allows the trainer to analyse the data at the end of a mornings work, when there is sufficient time. After identifying and applying the above objective methods to monitor training load in the young Thoroughbred racehorse, a simple and efficient subjective method of monitoring training load was created. The Equine Perceived Exertion (EPE) scale was established from Borg’s Rate of Perceived Exertion scale and Borg’s CR 1 – 10 scale. Six subjective parameters were used to assess each of seventeen Thoroughbred racehorses immediately following their return from sixty-five separate pre-training session. These parameters were behavioural characteristics, respiratory rate, perspiration, capillary refill time, oral mucous membrane colour, and muscle engorgement. Each parameter was assigned a scale and the sum of all parameters corresponded to an algorithm score. The algorithm scores then corresponded to a specific EPE score. Training load was calculated by multiplying the EPE score by the duration of the training session in minutes. There were weak positive correlations between algorithm score or EPE or training load and heart rate recovery 60-seconds post peak heart rate. It was concluded that monitoring training load in Thoroughbred racehorses in a pre-training program can be accomplished and needs to be multifaceted. The trainer must not rely on one particular method of monitoring training load but use both internal and external methods simultaneously to successfully monitor training load. Overall, this work has identified six methods to monitor the training load of Thoroughbred racehorses in pre-training. Further investigation is required to determine if these methods can be used to monitor the training load of Thoroughbred racehorses in gallop training, undertaking different training programs.

Q fever is a zoonotic disease with a significant impact on public health worldwide. It is caused by the highly infectious bacterium Coxiella burnetii. In humans, Q fever typically manifests as a self-limiting febrile syndrome. C. burnetii infection can also lead to chronic disease presentations and be life-threatening for individuals with underlying risk factors. Australia ranks among the countries with the highest Q fever report rates. Livestock species are the main source of infection for humans and dairy goats have been linked to some of the largest Q fever outbreaks recorded in recent times. In Victoria, Australia, a Q fever outbreak linked to a large dairy goat enterprise occurred in 2012 – 2014. This was the largest farm-associated Q fever outbreak recorded in the country. In the context of a steadily growing dairy goat sector, an improved understating of Q fever dynamics and the efficacy of available control strategies in dairy goats is required. This thesis documents a series of studies designed to better understand factors contributing to the spread of Q fever in this herd and the potential efficacy of different control measures. As a first step, we developed a software prototype for reporting herd performance in intensively managed dairy goat herds. This allowed productivity, reproductive performance, health and mortality to be documented, and enabled the performance of Q fever positive and Q fever negative does to be compared. With this system in place, a panel study was conducted to document the prevalence of C. burnetii shedding at the time of kidding. A heterogeneous C. burnetii shedding pattern was found, characterised by the presence of small numbers of ‘super shedder’ does. Further, does identified as ‘super shedders’ had reduced total lactation milk yields compared with C. burnetii negative does. Demographic data gathered using the herd health software prototype and data on Q fever prevalence from our panel study were used to develop a within-herd transmission model of Q fever. The model was used to assess the expected time to eradication by means of vaccination. In addition, the efficacy of segregation of pregnant does and culling of C. burnetii shedders at the time of parturition or abortion was assessed. Vaccination consistently led to disease eradication, although it required this intervention to be sustained for at least 6.5 (95% CrI: 4.1 to 11.3) years. A combination of vaccination with segregation of pregnant does or culling of C. burnetii shedders at parturition or abortion shortened the median time to eradication. The model was adapted to account for heterogenous shedding patterns and used to test the potential efficacy of a test and remove intervention. We found the removal of super shedders does alone would lead to Q fever eradication. However, tests with a high sensitivity for early detection of super shedders are required for this strategy to be effective.

While previous studies of Leptospira Hardjo in Victoria demonstrated a relatively high prevalence of exposure in both cattle (40%; Milner et al. 1980) and humans (22%; Sutherland 1988) these estimates need to be revised since they were made more than 30 years ago and leptospirosis vaccination programmes are now commonplace on Victorian dairy farms. This was a cross-sectional study to estimate the prevalence of Leptospira borgpetersenii sv Hardjo and Leptospira interrogans sv Pomona in dairy herds in South-Western Victoria. Fifty-three herds were enrolled into the study. Herd managers were asked to present 15 late-lactation cows that had fertility issues (cows that had not conceived or had delayed calving to conception intervals). Furosemide 500 mg was injected into the tail vein of eligible cows and a mid-stream urine sample of the second voiding collected to increase the likelihood of sampling leptospira .At the time of each herd visit a questionnaire was administered to herd managers asking them to provide details of methods used for controlling leptospirosis, including vaccination. Urine samples were pooled at the herd level and tested for leptospira spp. using qPCR. Pooled samples were then tested individually and samples that were positive, were tested for Leptospira Hardjo and Leptospira Pomona using qPCR. Three of the 53 pooled urine samples returned a positive result. The leptospira positive pools returned a minimum of three positive individual cow urine samples. Testing of individual cow urine samples identified an additional positive herd (with one weak positive and one inconclusive result), giving an apparent prevalence of approximately 8 (95% CI 1 to 13) leptospira-positive herds per 100 herds at risk. Based on the 53 completed questionnaires, leptospirosis vaccination programs were non-compliant with label directions in 35 out of 52 vaccinated herds: 67 (95% CI 54 to 78) out of 100 herds that routinely vaccinate for leptospirosis were doing so incorrectly. Of the 53 herds that took part in this study, only one herd was completely unvaccinated. Based on the findings from this study, and assuming the herds that took part in this study were an unbiased sample of the dairy herd population at risk, we estimate that close to one out of 10 dairy farms in South-Western Victoria are leptospirosis positive. While most herds are vaccinating for leptospirosis, most are doing so incorrectly. We conclude that herd managers need to be better educated regarding leptospirosis vaccination programs.

Castration and disbudding are necessary husbandry practices commonly performed in farms which have the potential to induce stress and pain if not managed correctly. Depending on the severity of a stressor, potential stress responses can suppress immune function, lower resistance to disease-causing factors, contribute to, and result in, reduced animal welfare. Scientific literature is discordant on what recommendations should be transferred to farmers and practitioners regarding the preferred age to castrate and disbud calves. This lack of scientific support to guide recommendations is even more noticeable in suckler beef calves, since most of the disbudding studies have used dairy calves and most of castration studies have used dairy calves or calves older than 6-mo-old. In this thesis, a series of studies were conducted to investigate: a) the age-related differences in plasma cortisol, acute phase protein, metabolite concentrations, haematological variables, behaviour and performance of suckler beef calves in response to Burdizzo castration; b) the effect of sex, breed and age on horn bud size of dairy and beef calves at disbudding; c) the relationship between age and horn bud size of dairy and beef calves at disbudding; d) the horn bud growth in Holstein Friesian calves. Results indicate that Burdizzo castration caused significant stress in 2.5- and 5.0-mo-old calves, characterized by the increase in plasma cortisol concentrations, scrotal swelling, and abnormal behaviours observed after treatment. These responses were reduced by castrating calves at younger age. A secondary stress response was observed, which was greater in older castrated calves, suggesting that calves castrated at older ages have a prolonged overall stress response, compared to younger calves. There was no effect of castration on haematology profiles, haptoglobin, metabolites, body temperature and growth performance. Thus, if calves are to be castrated without analgesia or anaesthesia, then it would be preferable to do this at 2.5-mo-old rather than later, such as 5.0-mo-old, in order to minimize the physiological stress, scrotal swelling and pain-related behaviours associated with Burdizzo castration. Horn bud size at time of disbudding, irrespective of calf sex, was greater in Holstein-Friesian calves than Charolais, Simmental and Limousin calves. These results suggest that horn bud develop differently in beef calves compared with dairy breed calves. In addition, the relationship between age and horn bud size was very weak in all breeds studied with no difference due to calf sex. This means that the age of the calf is not a good predictor of the horn bud size. Therefore, the age of the calves is not a good parameter to guide recommendations on the best age to disbud calves and use of pain relief due to the great variation in horn bud size within and between breeds. The findings of this thesis will have implications for the welfare of cattle in relation to best recommendations regarding the age at castration of suckler beef calves and disbudding of dairy and suckler beef calves.

Snake venom-induced consumption coagulopathy (VICC) is an important syndrome resulting from snake envenomation. The prothrombin activator notecarin D is responsible for this coagulopathy in tiger snake (Notechis scutatus) envenomation. In dogs, the presence of this coagulopathy is used to assist with the diagnosis of snake envenomation and, in some cases, can produce clinical signs associated with a coagulopathic envenoming. There are subtle differences in this syndrome between dogs and people, particularly concerning the development of clinical signs. Chapter two reviews tiger snake envenomation and the consequent VICC in Australian dogs with comparisons to human literature. The VICC syndrome between the two species share many clinical similarities; the key difference lies in the significance of clinical haemorrhage in people when compared with dogs. A brief overview of coagulation and coagulometry is provided. Chapter three is a prospective observational cohort study characterising the coagulation factor activity patterns in VICC from naturally occurring tiger snake envenomation in dogs. The changes in coagulation factors in a study cohort of tiger snake envenomed dogs were compared to a healthy control cohort. Fibrinogen, factor V and factor VIII were identified as the most consumed factors and displayed the fastest recovery to control values. Tests of coagulation times (i.e., prothrombin time and activated partial thromboplastin time) were similarly affected. Additionally, higher serum tiger snake venom concentrations were associated with greater reductions in factors II, V and VIII, and greater prolongations in both coagulation times. The elucidation of the consumption and recovery pattern of coagulation factors in dogs, contrasted with that of people, provides insight into the underlying mechanisms of tiger snake VICC and its treatment. This information can be used for further diagnostic and therapeutic endeavours.