DEVELOPMENT AND IMPROVEMENT OF CLINICAL TOOLS FOR REHABILITATING ENDANGERED BLACK COCKATOOS (CALYPTORHYNCHUS SPP.

Transcription

1 DEVELOPMENT AND IMPROVEMENT OF CLINICAL TOOLS FOR REHABILITATING ENDANGERED BLACK COCKATOOS (CALYPTORHYNCHUS SPP.) BACK TO THE WILD Anna T. Le Souëf BSc (Hons) BVMS This thesis is presented for the degree of Doctor of Philosophy Murdoch University, 2012

2 I declare that this thesis is my own account of my research and contains as its main content work which has not previously been submitted for a degree at any tertiary education institution. Anna T. Le Souëf 30 November 2012 ii

3 Hope is the thing with feathers that perches in the soul and sings the tune without the words and never stops at all - Emily Dickinson iii

4 ABSTRACT Black cockatoos (Calyptorhynchus spp.) in the biodiversity hotspot of southwest Western Australia are in decline, due to multiple threatening processes that include habitat loss, poaching, competition with other species, vehicle strikes and shootings. The combined efforts of the Perth Zoo Veterinary Department (PZVD) and the Western Australian Department of Environment and Conservation (DEC) have led to a specialised rehabilitation program centred upon the return of injured and debilitated black cockatoos to the wild. This PhD project focused on improving the efficacy of this program and expanding the current knowledge of the health status and biology of black cockatoos. The medical records of 565 black cockatoos that were admitted to the PZVD from 2000 to 2009 were analysed to determine the effect of the birds clinical presentation on survival. Anaemia, superficial and deep soft tissue injuries, fractures, paralysis or paresis and abnormal faecal cytology were identified as significant factors when determining the likelihood of survival of cockatoos undergoing the rehabilitation process. In addition, data on the life histories and origin of admitted cockatoos, post-mortem examination results and post-release information is presented. Haematologic and serum biochemical reference values are also presented for the three species of black cockatoos that are regularly admitted to the PZVD. iv

5 The significance of Chlamydia psittaci infection to the black cockatoo rehabilitation program was investigated. No wild birds admitted to the PZVD were found to be positive for C. psittaci infection, which suggests that this infection either is not present in wild populations, or has a low prevalence. However, the study found that black cockatoos undergoing rehabilitation are at risk of contracting the disease during their stay in captivity through contact with infected birds. These results highlight the importance of disease surveillance to the management of an avian rehabilitation program. The thesis includes the results of an aviary trial to investigate methods of attaching transmitters to black cockatoos. Mean retention times for the transmitter packages ranged from 44 to 384 days. These results support the potential feasibility of using transmitters on cockatoos post-release, with sufficient retention times to allow for the collection of valuable movement and survival data. Among the gaps in current knowledge of wild black cockatoos, perhaps the most concerning is that regarding the age structure of wild populations. The thesis encompasses a study that established an aging tool for black cockatoos, based on pentosidine analysis from the skin of 53 black cockatoos of known age. The result is a vital first step towards understanding the population dynamics of wild black cockatoos, and will help provide further information about the life histories of cockatoos admitted to the PZVD following injury or debilitation. v

6 ACKNOWLEDGEMENTS I would first like to thank my supervisors. My principal supervisor, Kris Warren has been an incredible source of support and inspiration for many years. I have been so grateful for her endless enthusiasm for the project and for guiding me on my first steps on the pathway to a career in conservation medicine. Simone Vitali was always available to me for advice and was a great source of practical and innovative ideas, particularly with respect to the clinical aspects of the project. I am grateful to Carly Holyoake for her help with the epidemiological aspects of my project; Carly s support and friendship during this time will not be forgotten. Stan Fenwick is also thanked gratefully for his input, and for his belief in the project from the early stages. I would like to extend a thank you to DEC for their cooperation with different aspects of the study. It would not have been possible without their support; in particular, Rick Dawson, who was instrumental in assisting with sample collection and fieldwork, always with a huge amount of enthusiasm and humour. Many thanks also go to Birdlife Australia and WWF-Australia for their help with funding the transmitter aviary trial. Thanks to Crissa Cooey and Professor Hillar Klandorf at the Division of Forestry and Natural Resources at West Virginia University for their helpful collaboration on the aging study and for kindly hosting me during my stay in the US. vi

7 I am enormously appreciative to my colleagues at the PZVD for sample collection and support. I would also like to acknowledge Glenn and Andrea Dewhurst and the team at the Kaarakin Black Cockatoo Rehabilitation Centre, where I have spent many hours and always felt welcome. If only those beautiful birds, such as Lenny, could also realise how they have helped their species by being a part of this project. To my wonderful family: Mum, Dad, Tim and Kate who have provided lots of support and love over the years. And finally to my husband and best friend Hugh, for being behind me all the way and supporting me through all the ups and downs. vii

8 PUBLICATIONS Le Souëf A, Holyoake C, Vitali S, Warren K Haematologic and serum biochemical reference values for three species of Western Australian black cockatoos (Calyptorhynchus spp.). Journal of Avian Medicine and Surgery. 27(1): Le Souëf A, Stojanovic D, Burbidge A, Dawson R, Heinsohn R, Vitali S, Warren K Retention of transmitter attachments on black cockatoos (Calyptorhynchus spp.). Pacific Conservation Biology. 19(1): viii

9 CONFERENCE AND COMMUNITY PRESENTATIONS The development of conservation and prognostic indicators for the rehabilitation of black cockatoos (Calyptorhynchus spp.). WDA Annual conference of the Wildlife Disease Association (Australasian Section). Dryandra Woodland, Western Australia September Epidemiological study of black cockatoos (Calyptorhynchus spp.) admitted to the PZVD. WDA Annual conference of the Wildlife Disease Association (Australasian Section). Kioloa, New South Wales September Veterinary rehabilitation of black cockatoos at the PZVD. Carnaby s Black cockatoo Symposium. Department of Environment and Conservation, Kensington, Western Australia. 1 December From pest to plight: Western Australia s black cockatoos. West Virginia University Wildlife Society meeting. Morgantown, West Virginia, USA. 2 September A study of endangered black cockatoos (Calyptorhynchus spp.) admitted to the PZVD (Poster). 12 th Conference of the International Society for Veterinary Epidemiology and Economics. Durban, South Africa August ix

10 MEDIA REPORTS Key to improving cocky treatment and rehab. Ecos. CSIRO Publishing, Collingwood. 149: 6. Contos, F Saving the Carnaby s cockatoo. Explore magazine. Murdoch University, Perth. 5(5): 5. Foster, B On the wing. Fremantle Herald. 30 May. Glitsos, L Backpack trackers let researchers collar cockies. Science Network WA. (Accessed 3 June 2009). Available from: au/topics/environment/192-news/2607-backpack-trackers-let-researchers-collarcockies-. Lau, J Carnaby s cockatoo: tracking black cockatoos. The Bird Observer. Bird Observation and Conservation Australia, Nunawading. 861: 18. Le Souëf, A Cockies get wired! Working towards tracking black cockatoos. Cocky Notes. Birds Australia, Floreat. 14: 3. Rice, D Polly want a tracker? Intouch Alumni Magazine. Murdoch University, Perth. Spring: 16. Scarparolo, D Tracking cockatoos. Newspaws. Perth, Perth Zoo. Winter: 6. AWARDS Dean s Prize - Best in Show: A Tool for the Age Estimation of Black cockatoos (Calyptorhynchus spp.). School of Veterinary and Biomedical Sciences, Murdoch University Research Poster Day x

11 TABLE OF CONTENTS Declaration Abstract Acknowledgements Publications Conference and Community Presentations Media Reports Awards Table of Contents List of Tables List of Figures Glossary of Abbreviations and Acronyms ii iv vi viii ix x x xi xvii xix xxiii CHAPTER 1 GENERAL INTRODUCTION 1.1 Introduction History and taxonomic relationships of south-west Western Australian black cockatoos White-tailed black cockatoos Red-tailed black cockatoos Biology Carnaby s cockatoo Baudin s cockatoo Red-tailed black cockatoo Conservation status, threatening factors and recovery efforts Carnaby s cockatoo Baudin s cockatoo Red-tailed black cockatoo Rehabilitation of wildlife as a conservation strategy Rehabilitation of black cockatoos at Perth Zoo Veterinary Department Conclusions Aims Structure of chapters 37 xi

12 CHAPTER 2 AN EPIDEMIOLOGICAL STUDY OF WILD BLACK COCKATOOS ADMITTED TO THE PERTH ZOO VETERINARY DEPARTMENT FROM Introduction Retrospective cohort studies Published studies on avian rehabilitation programs Importance of this study to the veterinary management of wild black cockatoos at PZVD Aims Materials and methods PZVD black cockatoo treatment protocol Sex determination using endoscopy and DNA methods Acquisition of data Data management Statistical analysis Spatial distribution of cockatoo rescue locations Post-release data Analysis of post-mortem examination records Results Admission trends Signalment Temporal patterns in admissions Clinical presentation Time spent in care Reasons for euthanasia Outcomes Univariate analysis of risk factors against outcome Multivariate analysis of risk factors against outcome 79 xii

13 2.4.4 Univariate analysis of risk factors against outcome among cockatoos with fractures Spatial distribution of location found Data from cockatoos that were released Post-release data Post-mortem examination results and evidence of disease Mycotic pneumonia (aspergillosis) Parasites Gastrointestinal parasites Ectoparasites Discussion Admission trends Analysis of factors affecting outcome Spatial distribution of locations at which cockatoos were found Post-release data Post-mortem findings and evidence of disease Recognition of limitations of the study Conclusions 112 CHAPTER 3 A STUDY OF CHLAMYDIA PSITTACI INFECTION IN WILD BLACK COCKATOOS PRESENTED TO THE PZVD FOR TREATMENT AND REHABILITATION 3.1 Introduction Developmental cycle Transmission Pathogenesis Clinical disease and pathology in psittacine birds Distribution and host range Significance in wild populations Public health concerns Diagnosis 128 xiii

14 Isolation in cell culture Polymerase chain reaction (PCR) Serology Immunoassays for antigen detection Immunohistochemistry and histochemical staining DNA microarray technology Treatment and control Aims Materials and methods Serological testing ImmunoComb Avian C. psittaci Antibody Test Kit Sample collection Sample testing using the Immunocomb Kit PCR testing Sample collection DNA extraction, PCR and DNA sequencing Statistical analysis Results Discussion Results of testing Historical cases of chlamydiosis among wild black cockatoos admitted to the PZVD Outbreak of chlamydiosis at the Kaarakin Black Cockatoo Rehabilitation Centre Management recommendations Conclusions 156 xiv

15 CHAPTER 4 HAEMATOLOGIC AND PLASMA BIOCHEMICAL REFERENCE VALUES FOR THREE SPECIES OF BLACK COCKATOOS (CALYPTORHYNCHUS SPP.) 4.1 Introduction Materials and methods Results Discussion 168 CHAPTER 5 TRANSMITTER ATTACHMENTS FOR LARGE PSITTACINES: BLACK COCKATOOS (CALYPTORHYNCHUS SPP.) AS A CASE STUDY 5.1 Introduction Methods Study area and species Transmitter attachment Statistical analysis Results Discussion 182 CHAPTER 6 THE DEVELOPMENT OF AN AGING CURVE FOR BLACK COCKATOOS USING PENTOSIDINE ANALYSIS 6.1 Introduction Aging in birds Longevity in black cockatoos Age estimation and cockatoo conservation Methods of age determination in birds Pentosidine measurement Aim Materials and methods Sample procurement Sample collection Skin sample preparation Hydroxyproline analysis Pentosidine analysis 197 xv

16 6.3.6 Statistical analysis Results Discussion Conclusions 201 CHAPTER 7 GENERAL DISCUSSION 7.1 Retrospective studies can provide important information for conservation management The value of rehabilitating endangered birds A review of current biosecurity procedures at the Rehabilitation Centre and further recommendations The role of avian rehabilitation programs in disease surveillance Public health significance of Chlamydia psittaci in black cockatoos Conclusions and future research Post-release monitoring of rehabilitated black cockatoos Estimation of ages for wild black cockatoos 217 REFERENCES 219 APPENDICES 1a Excerpt from database used to analyse medical records of black cockatoos 272 1b Key to database used to analyse medical records of black cockatoos List of black cockatoos used in the final curve for relationship between pentosidine and age (after removal of outliers) 275 xvi

17 LIST OF TABLES Table 2.1 Retrospective studies on the rehabilitation of wild birds. 41 Table 2.2 Frequency of fractures according to location in black cockatoos presented to the PZVD from Table 2.3 Days spent in the PZVD for various clinical signs for surviving black cockatoos that presented from Table 2.4 Factors analysed against outcome for black cockatoos presented to the PZVD from Table 2.5 Significant variables from multivariate regression analysis of black cockatoos, excluding birds euthanased on the first day of presentation, for (survival as dependent variable). 79 Table 2.6 Factors analysed against outcome for black cockatoos presenting with fractures (excluding birds euthanased on the first day). 80 Table 2.7 Post-mortem findings for 161 black cockatoos examined at the PZVD from Table 2.8 Characteristics of mycotic pneumonia found at necropsy in 13 black cockatoos from Table 3.1 Known hosts of C. psittaci serovars (Geens et al. 2005a; Anderson and Christian Franson 2007). 117 Table 3.2 Number of cockatoos tested for C. psittaci infection during various stages of rehabilitation. 139 xvii

18 Table 3.3 Results of C. psittaci diagnostic testing using serology for antibody detection and PCR for antigen detection for black cockatoos during different stages of rehabilitation. 145 Table 4.1 Origins of the black cockatoos sampled for compilation of haematologic and biochemical reference ranges. 160 Table 4.2 Values for haematologic and biochemical tests for Carnaby s cockatoos. 164 Table 4.3 Values for haematologic and biochemical tests for Baudin s cockatoos. 165 Table 4.4 Values for haematologic and biochemical tests for forest redtailed black cockatoos. 166 Table 4.5 Values for analytes that differed significantly between sampling method (jugular vein under anaesthesia vs. metatarsal vein under manual restraint) for Carnaby s cockatoos. 167 Table 5.1 Number of black cockatoos of each species assigned to different transmitter groups. 174 Table 5.2 Transmitter retention times (days) for collar, 12g tail-mount, 6g tail-mount, 12g harness, 16g harness and 20g harness transmitter packages on black cockatoos. 181 Table 6.1 Origins of the black cockatoos from which skin samples were taken for pentosidine analysis. 193 Table 6.2 Ages of black cockatoos from which skin samples were taken for pentosidine analysis. 194 xviii

19 LIST OF FIGURES Figure 1.1 Comparison of bill and ear covert colour of female (left) and male Carnaby s cockatoos. 6 Figure 1.2 Distribution of the Carnaby s cockatoo (Johnstone et al. 1998). 7 Figure 1.3 Heads of the Carnaby s cockatoo (A) and Baudin s cockatoo (B), showing differences in upper bill shape and length (Johnstone et al. 1998). 13 Figure 1.4 Distribution of the Baudin s cockatoo (Johnstone et al. 1998). 14 Figure 1.5 Distribution of the forest and inland red-tailed black cockatoo in Western Australia (Johnstone et al. 1998). 18 Figure 2.1 Appearance of gonads in black cockatoos at various stages of development, as seen via endoscopy (a: immature or quiescent testis (black arrow); b: immature or quiescent ovary (black arrow); c: mature active testis; d: mature active ovary). 53 Figure 2.2 Body condition scoring system for black cockatoos using crosssectional size of pectoral musculature (Perth Zoo). 56 Figure 2.3 Internal haemorrhage associated with traumatic injury in a black cockatoo (white arrows show free blood clots in the anterior coelomic cavity). 63 Figure 2.4 Immature male Baudin s cockatoo showing female physical characteristics (sex determined by DNA testing and visualisation of immature testes). 64 Figure 2.5 Seasonal variation in gonadal appearance among black cockatoos that were endoscopically sexed. 65 xix

20 Figure 2.6 Number of black cockatoos admitted to the PZVD from Figure 2.7 Seasonality of the numbers of black cockatoo species admitted to the PZVD, pooled for 10 years, from Figure 2.8 Frequency of black cockatoos admitted to the PZVD according to season from Figure 2.9 Relative frequency of reason for admission for black cockatoos treated at PZVD during Figure 2.10 Relative frequency of clinical signs for black cockatoos presented to the PZVD from Figure 2.11 Primary clinical signs seen in black cockatoos euthanased on first day of presentation at the PZVD from Figure 2.12 Outcomes for black cockatoos treated from Figure 2.13 Outcomes for black cockatoos presented to the PZVD with fractures from Figure 2.14 Map showing urban areas in Perth, Western Australia in which black cockatoos were found before presentation to the PZVD from Figure 2.15 Map showing regional areas in south-western Western Australia in which black cockatoos were found before presentation to the PZVD from Figure 2.16 Necropsy lesion associated with aspergillosis in a black cockatoo (green arrow shows white-green caseous nodule in the left lung lobe). 90 xx

21 Figure 2.17 Percentage of aspergillosis cases among black cockatoos presented to the PZVD, according to year, from Figure 3.1 Developmental cycle of chlamydiae (adapted from Vanrompay et al. 1995a). 118 Figure 3.2 Swabbing the conjunctiva of a Carnaby s cockatoo for C. psittaci PCR testing. 143 Figure 3.3 Carnaby s cockatoo (Perth Zoo record number B ) diagnosed with chlamydiosis showing splenomegaly (white arrow) and hepatomegaly (green arrow). 151 Figure 4.1 A normal eosinophil and erythrocytes in the peripheral blood film from a parrot (Wright s stain, 1000X) (Campbell and Ellis 2007). 171 Figure 5.1 Field anaesthesia set up for transmitter attachment. 176 Figure 5.2 Harness design for black cockatoos showing modified Xattachment method (and the central cross rests on the sternum of the bird). 177 Figure 5.3 6g tail-mounted PTT attached to ventral tail feather. 177 Figure 5.4 Collar attached to cockatoo with barrel resting above the sternum. 179 Figure 6.1 Collection of the 6mm skin biopsy from the inner patagium of a black cockatoo (a: injecting lignocaine; b: making the biopsy incision; c: sealing the incision with tissue adhesive; d: the final appearance). 195 xxi

22 Figure 6.2 Pentosidine standard curve used to calculate pentosidine concentrations in the skin of black cockatoos. 197 Figure 6.3 Pentosidine (Ps) concentration in the skin as a function of chronological age in black cockatoos (n = 49). 199 xxii

23 GLOSSARY OF ABBREVIATIONS AND ACRONYMS AGE advanced glycation end-product ANGIS Australian National Genomic Information Service AST aspartate aminotransferase BAWA Birds Australia (Western Australia) BFDV psittacine beak and feather disease virus BOHB beta-hydroxybutyrate C Celsius Ca calcium CF complement fixation CHD chromo-helicase-dna CI confidence interval CK creatinine phosphokinase Cl chloride cm centimetre DAFWA Department of Food and Agriculture Western Australia DEC Department of Environment and Conservation xxiii

24 DNA deoxyribonucleic acid ELISA enzyme-linked immunosorbent assay G gauge g gram GA general anaesthesia h hour Hb haemoglobin HRM high resolution melt (curve analysis) IFAT immunofluorescent antibody testing IgG immunoglobulin G IgM immunoglobulin M IHC immunohistochemical IUCN International Union for Conservation of Nature K potassium kg kilogram km kilometre L litre LPS lipopolysaccharide xxiv

25 m metre MCH mean corpuscular haemoglobin MCHC mean corpuscular hemoglobin concentration MCV mean corpuscular volume MedARKS Medical Animal Records Keeping System mg milligram MLVA Multiple Loci Variable Number of Tandem Repeats (VNTR) Analysis min minute ml millilitre MOMP major outer membrane protein MZN modified Ziehl-Neelsen n number Na sodium nm nanometre OR odds ratio PZVD Perth Zoo Veterinary Department P phosphorus xxv

26 P probability value PBFD psittacine beak and feather disease PCR polymerase chain reaction PCV packed cell volume PO per oral Ps pentosidine concentration PTT platform transmitter terminal RBC red blood cell rpm revolutions per minute s seconds TPP total plasma protein UK United Kingdom USA United States of America VNTR Variable Number of Tandem Repeats WCC or WBC white cell count WWF World Wide Fund for Nature μl microlitre xxvi

27 CHAPTER ONE: GENERAL INTRODUCTION 1

28 1.1 INTRODUCTION There are three species of black cockatoos endemic to south-west Western Australia: Carnaby s cockatoo (Calyptorhynchus latirostris) (Carnaby 1948), Baudin s cockatoo (Calyptorhynchus baudinii) (Lear and Ackermann 1832) and the forest red-tailed black cockatoo (Calyptorhynchus banksii naso) (Gould 1865). The wild populations of these species are declining as a result of many factors including habitat loss and fragmentation, poaching, competition for nesting resources with other species, vehicle strike and shooting (Saunders 1990; Mawson 1997; Mawson and Johnstone 1997; Saunders and Ingram 1998). The ranges of distribution of each of the species have been reduced significantly during the past half-century, coinciding with large scale land-clearing which occurred in Western Australia during that period (Mawson and Johnstone 1997). All three species are recognised as either endangered or threatened at either state or federal levels (Government of Western Australia 1950; Australian Government 1999). As part of recovery efforts attempting to stabilise free-ranging populations, wild injured or debilitated black cockatoos from urban and regional Western Australia are treated and rehabilitated for release. This rehabilitation program has been in progress since 2000, when the Department of Environment and Conservation (DEC) (formerly the Department of Conservation and Land Management) and Perth Zoo recognised a need for intensive black cockatoo rehabilitation efforts due to apparent significant losses from vehicle strikes and shootings. The most common presenting problem is vehicle strike, however shot injuries are also frequently encountered. 2

29 The author worked as a clinical veterinarian at the Perth Zoo Veterinary Department (PZVD) for the duration of this PhD project and was involved in every aspect of the veterinary treatment of black cockatoos. 1.2 HISTORY AND TAXONOMIC RELATIONSHIPS OF SOUTH- WEST WESTERN AUSTRALIAN BLACK COCKATOOS White-tailed black cockatoos Historically there has been confusion regarding the taxonomic separation of the white-tailed black cockatoo. A white-tailed black cockatoo was described and depicted by ornithological illustrator Edward Lear as Calyptorhynchus baudinii in 1832 (Lear and Ackermann 1832), named after the French explorer Nicolas Baudin. Mathews (1912) recorded two subspecies of white-tailed black cockatoo, one with a long narrow bill and the other with a shorter wider bill. He incorrectly assumed Lear s figure of baudinii to be short-billed, describing the other subspecies, tenuirostris, as having a much longer and narrower bill. Lear s specimen is in the American Museum of Natural History and is most likely a baudinii. Mathews (1917) later changed his mind and stated that the very restricted range of this bird is conducive to no subspecific forms. The Western Australian naturalist Ivan Carnaby considered that the white-tailed black cockatoo in the Lake Grace district of Western Australia was a subspecies of C. baudinii, and later formally named it C. b. latirostris (Carnaby 1948). Serventy and Whittell (1967) accepted the two subspecies, although their nomenclature was incorrect. By the 1960s it was conventional to refer to the two forms of white-tailed black cockatoo as being two subspecies, although one author 3

30 (Forshaw 1969, 1973) described C. baudinii as a subspecies of the yellow-tailed black cockatoo C. funereus. In 1974 Saunders (1974b) concluded that, due to distinct differences in culmen length, wing length, skull size, distribution and diet, the two forms of the whitetailed black cockatoo were separate species, describing them as C. baudinii latirostris and C. b. baudinii. This trend continues to the current period, although they are now known as Carnaby s cockatoo and Baudin s cockatoo (C. latirostris and C. baudinii). Previous names for these cockatoos include the white-tailed black cockatoo, or the short- and long-billed white-tailed black cockatoo (Johnstone et al. 1998). Recent studies have determined genetic similarity between the white-tailed black cockatoo species and a closer examination of the taxonomic relationship between these species may be warranted (White et al. 2011). Many historical observational records and even some recent literature describing white-tailed black cockatoos in Western Australia do not distinguish between the Carnaby s cockatoo and the Baudin s cockatoo, and it is not always possible to determine which species records refer to (Johnstone and Kirkby 2008) Red-tailed black cockatoos There are five subspecies of red-tailed black cockatoo (Calyptorhynchus banksii) in Australia. The red-tailed black cockatoo was originally named Psittacus banksii by John Latham (1790) in commemoration of the English botanist Sir Joseph Banks. The south-west subspecies Calyptorhynchus banksii naso was first described by John Gould in 1837 (Gould 1865). The word naso means nose, in reference to the large size of the bill relative to other red-tailed black cockatoos (Ford 1980). Ford (1980) supported the recognition of a south-west subspecies of 4

31 the red-tailed black cockatoo, the forest red-tailed black cockatoo (Calyptorhynchus banksii naso). The inland red-tailed black cockatoo (Calyptorhynchus banksii samueli) was first described by Mathews in BIOLOGY Carnaby s cockatoo Physical description The Carnaby s cockatoo is a large cockatoo with a recorded body length of 53-58cm and a weight of g (Johnstone et al. 1998). The adult male has brownish black plumage and the feathers are tipped with dusky white, giving a scalloped appearance. The feathering on the head is a glossy black with dusky or brownish white ear coverts (Johnstone et al. 1998). The central tail feathers are mostly black, however the lateral tail feathers on the ventral side are mostly white, with a black base and a broad black tip. Male Carnaby s cockatoos that have reached the age of sexual maturity (approximately four years old) have a pink peri-ophthalmic ring which reddens during the breeding season. The bill is black or greyish black, and becomes glossier in the breeding season (R. Johnstone, personal communication, 5 June 2008). The female Carnaby s cockatoo is similar in appearance to the male, but differs in having yellowish white ear coverts, often with a more distinct rostral margin, a greyish or bone-coloured bill with a black tip and a dark grey or grey eye rim (Johnstone et al. 1998) (Figure 1.1). 5

32 Figure 1.1 Comparison of bill and ear covert colour of female (left) and male Carnaby s cockatoos. The appearance of juvenile Carnaby s cockatoos is generally similar to the adults (i.e. males with dull ear coverts and females with brighter ear coverts), however the bill remains pale until the second year of life, when in males it often begins to darken (Johnstone et al. 1998). The sex of juvenile Carnaby s cockatoos from the age of nestlings can generally be estimated by the appearance of the ear coverts, but in some individuals the physical signs can be ambiguous (R. Johnstone, personal communication, 5 June 2008). Sexing using physical characteristics is only reliable once the cockatoo has reached sexual maturity at approximately four years of age. Ecology and behaviour The Carnaby s cockatoo occurs only in the south-west region of Western Australia (Saunders 1982) (Figure 1.2) and breeds in woodland and shrubland, moving to coastal areas in flocks during the non-breeding season (BirdLife International 2010b). Some flocks may remain in the same high rainfall area year round to breed and forage, while those that are in drier parts of the range move to higher rainfall areas for breeding (Marchant et al. 1990). The Carnaby s cockatoo 6

33 is uncommon to common in the subhumid zone and wetter parts of the semiarid zone, and has a scarce patchy distribution in drier parts of the range and in the deep south-west (Johnstone et al. 1998). Figure 1.2 Distribution of the Carnaby s cockatoo (Johnstone et al. 1998). Since the widespread clearing of the wheatbelt in the past half-century, the habitat range has shifted west- and southwards. Carnaby s cockatoos are seen in pairs, small flocks or large flocks of up to 2,000 in the non-breeding season (Johnstone et al. 1998). The call of the Carnaby s cockatoo is a distinctive wailing wy-lah or wee-looo, repeated and often ending with an upward inflection (Marchant et al. 1990; 7

34 Johnstone and Kirkby 2008), with variations between individuals, between sexes and between birds from different regions. The majority of calls appear to be related to group maintenance and co-ordination (Saunders 1983). The scarcity of vocalisations during Carnaby s cockatoo sexual encounters is likely related to the fact that the birds are long-lived and pair for life. Individuals can be identified by their calls by discriminating the lengths of parts of the call (Saunders 1983). Carnaby s cockatoos are largely arboreal but may be seen feeding on the ground on fallen fruits and seeds (Marchant et al. 1990). They characteristically break open the rim of fruit (e.g. from marri (Carymbia calophylla)), unlike the Baudin s cockatoo, which is able to extract the seeds directly from the fruit with its longer upper beak (Saunders 1974b). They have been known to feed on native vegetation such as the seeds of Hakea, Grevillea, Dryandra, Eucalyptus, Allocasuarina, Banksia spp. and double-gees (Emex australis), and the flowers of Dryandra, Lambertia, Banksia, Eucalyptus, Grevillea and Calistemon spp., as well as some species of introduced vegetation (Saunders 1974a, 1977a, 1980; Johnstone et al. 1998) and agricultural crops such as canola (Jackson 2009). Carnaby s cockatoos have also been observed feeding on the nectar of native flowers (Johnstone et al. 1998). Carnaby s cockatoos can be described as omnivorous, as they also feed on insects hidden in the flowers or fruits of banksia species, such as the wood-boring moth (Arthrophora sp.) found in the Lesueur banksia (Banksia tricuspis) (van Leeuwen and Lamont 1996), the seed-eating weevil found in Banksia attenuata (Scott and Black 1981) and insect larvae from the families Cerambycidae and Pyralidae (Saunders 1980). Carnaby s cockatoos may control insects in forest ecosystems at 8

35 a local level (Scott and Black 1981) and due to their ingestion of seed-eating weevils are thought to have a net positive effect on seed production of the rare Lesueur banksia (Lamont and van Leeuwen 1988; Lamont et al. 2007). Most black cockatoos reach sexual maturity at about four years of age, however the Carnaby s cockatoo (in particular the females) may not reach reproductive age until they are six or seven years old (Cameron 2007). Adults prefer to pair for life and usually remain paired throughout the year, including the non-breeding season (Saunders 1977a). Pairs are seldom apart except for when the female is in the nest and re-pairing usually occurs only after the loss of a mate (Saunders 1979b; Marchant et al. 1990). Carnaby s cockatoos may lay their clutches any time between July and December, although most lay from July to September. Breeding mainly occurs in the semiarid and subhumid interior, in any eucalypt with a hollow of suitable size, such as wandoo (Eucalyptus wandoo), salmon gum (Eucalyptus salmonophloia), red morrell (Eucalyptus longicornis), York gum (Eucalyptus loxophleba), marri and tuart (Eucalyptus gomphocephala) (Saunders 1979b; Johnstone et al. 1998). Carnaby s cockatoos nest in tree hollows, which are formed as a result of termites and fungi attacking the heartwood of live or dead standing trees. They will nest in any species of large mature eucalypt which has a hollow of a suitable size (Saunders 1979a). The lowest average age recorded for any species of tree with a nest hollow suitable for parrots is 275 years (Mawson and Long 1994). It is common for the Carnaby s cockatoo to lay two eggs, however the second nestling usually dies within 48 hours of hatching (Saunders 1982b). The number of eggs laid may be an adaptation to the largest number of young for which parents can 9

36 provide enough food and also insures against the loss of one chick during incubation (Lack 1954; Saunders 1982b). The difference in hatching dates between the two eggs of four to seven days can result in large size differences between clutchmates, and the larger chick may take up the most advantageous position in the hollow as well as having a physical advantage (Saunders 1982b). Incubation takes 29 days and the nesting period is approximately days. Only the female incubates and broods, and for the first days after hatching the chick will be fed exclusively by the hen, after which both parents may feed the chick (Saunders 1979b). In years when food is abundant, two young may be raised, as foraging birds have a high return per unit effort (Saunders 1982b). When this occurs, the younger nestling fledges a week or two after the sibling and is fed while the parents and other nestling remain in the vicinity of the nest tree. The chick fledges around ten or eleven weeks after hatching, with recorded weights between 351 and 750g (Saunders 1982b). After fledging, the chick is usually fed by the parents for about three months, although this can be up to six months (Cameron 2007). The young bird will then remain associated with its parents at least until the next breeding season, or up to several years. This long association ensures that the young learn about the type of foods that may be eaten, how to forage and what range to forage over (Saunders 1982b). Juveniles often do not follow the adults to the breeding areas, but there is little known about movements of young birds during the breeding season (Saunders 1977a). 10

37 Nesting may occur in several hollows in the same tree. However, female cockatoos take two to three weeks to select and prepare a hollow, during which time they attempt to limit conspecifics from inhabiting nearby hollows. After this time the female will no longer interact with other females searching for hollows. This exclusion of birds from unused nest hollows limits the total number available, as hollows in the defended area are unusable for about 15% of the breeding season. Therefore the overall availability of hollows does not necessarily limit the population in itself, however the breeding behaviour of female cockatoos may exert some limiting effect on the population (Saunders 1979a). Carnaby s cockatoos generally forage close to nesting areas, usually within one or two kilometres, however this depends on the availability of suitable foraging vegetation (Saunders 1977a). A study by Saunders (1977a) indicated that discontinuous verges and isolated patches of vegetation are not sufficient as sources of food, as the cockatoos prefer to follow vegetation corridors leading from the main reserve. Mature adult birds spend the non-breeding season in foraging flocks that may move some distance from the breeding areas. At the start of the following breeding season the adults return to the same breeding areas but the immature cockatoos remain away from the breeding areas. Little is known about the movements of immature birds until they begin breeding at about four years of age (Saunders 1977a). Once breeding has finished, fledglings remain with their parents and band together into foraging flocks. 11

38 1.3.2 Baudin s cockatoo Physical description The Baudin s cockatoo has a very similar appearance to the Carnaby s cockatoo, with a length of 50-56cm and a recorded weight range of g (Johnstone et al. 1998). The morphological differences between the Carnaby s and Baudin s cockatoo have been well documented. The main difference between the two species is the length of the culmina (the ridge along the upper bill) (Figure 1.3); the mean length of the culmen of adult C. baudinii is 53mm while adults of C. latirostris have a mean culmen length of 44.2mm (Saunders 1979c). The length of the rhamphotheca (horny covering of the maxilla) may be assumed to have been subject to change according to differing diets rather than evolutionary adaptation; however the two species also display distinct differences in the length of the bone of the underlying maxillae (Saunders 1979c). 12

39 Figure 1.3 Heads of the Carnaby s cockatoo (A) and Baudin s cockatoo (B), showing differences in upper bill shape and length (Johnstone et al. 1998). Ecology and behaviour There are differences in the flight or contact calls between the two white-tailed black cockatoo species; an experienced listener can differentiate between the species on the call alone (Saunders 1979c, 1983). The call of the Baudin s cockatoo tends to be shorter and disyllabic compared with the Carnaby s cockatoo (Marchant et al. 1990). Baudin s cockatoo feeds mainly on eucalypt forests, however are also attracted to seeding marri, Banksia, Hakea and Erodium spp. and to fruiting apples and pears 13

40 (Ashby and Le Souëf 1928; Saunders 1974a; Johnstone et al. 1998). It will also strip bark from dead trees to search for beetle larvae (Johnstone et al. 1998). Baudin s cockatoos and Carnaby s cockatoos have been seen feeding in mixed groups (Marchant et al. 1990). The Baudin s cockatoo occurs exclusively in south-west Western Australia with a habitat range of 2000 km 2 (Garnett and Crowley 2000) (Fig. 1.4) and is mainly found in the humid and subhumid zones (areas with average rainfall of 600mm or more) (Johnstone and Kirkby 2008). Its range is similar to the forest red-tailed black cockatoo. Figure 1.4 Distribution of the Baudin s cockatoo (Johnstone et al. 1998). 14

41 Depending on their region of origin, Baudin s cockatoo is either a resident, a nomad or migrant, with most of the population tending to vacate the colder parts of the range in autumn to migrate northwards or towards the coast (Johnstone and Kirkby 2008). During the non-breeding season, it forages widely, with its range overlapping that of the Carnaby s cockatoo. Baudin s cockatoo is gregarious and moves in flocks, with aggregations of up to 1,200 birds recorded at drinking sites or roosts (Johnstone and Kirkby 2008). In addition to the relative paucity of knowledge on the distribution of the Baudin s cockatoo, very little published information exists on their breeding behaviour and biology, although it is thought to be similar to the Carnaby s cockatoo. Those individuals that migrate during the year move towards non-breeding traditional roosts in the central and northern parts of the Darling Scarp (from Collie ( S E) to Mundaring ( S E)) in late summer. There appears to be a shift westward to the southern Swan Coastal Plain, just prior to flocks moving south in mid-august for the breeding season (Johnstone and Kirkby 2008). Baudin s cockatoo form strong pair bonds and probably mate for life (Johnstone and Kirkby 2008). It usually nests in the hollows of karri (Eucalyptus diversicolor), marri and wandoo trees (Johnstone et al. 1998) and nests have been seen to be used for at least two consecutive years (Ashby and Le Souëf 1928). Egg laying has been recorded in the months from August to December, however pairs have also been seen prospecting for hollows in most months and outside the breeding range (Johnstone and Kirkby 2008). The annual reproductive rate of the 15

42 Baudin s cockatoo is only 0.6 chicks per pair, probably precluding it from replacing the large numbers currently shot by orchardists (Johnstone et al. 1998). Although the generalised distribution of the Baudin s cockatoo is known, more detailed information regarding current distribution and critical habitat areas is lacking. This is partly because early accounts of white-tailed black cockatoos did not distinguish between the Carnaby s cockatoo and the Baudin s cockatoo. The flocks and nests of the Baudin s cockatoo are also difficult to locate as they are in the canopies of dense forest and in tall trees (Chapman 2008) Red-tailed black cockatoo There are five subspecies of red-tailed black cockatoo (C. banksii). Only the forest red-tailed black cockatoo (C. b. naso) and the inland red-tailed black cockatoo (C. b. samueli) are found in south-west Western Australia. Physical description The forest red-tailed black cockatoo is a physically spectacular cockatoo with a length of 53-55cm and a recorded weight range of g (Johnstone et al. 1998). It has a large mandible and maxilla presumably an adaptation to feeding on marri nuts, from which it chews out the base of the mature seed capsule (Saunders et al. 1985). The male forest red-tailed black cockatoo has mainly glossy black plumage, with a short rounded crest. The central tail feathers are glossy black, while the lateral tail feathers are glossy black with a broad central band of bright orange-red. The peri-ophthalmic ring and bill are dark grey. 16

43 The female forest red-tailed black cockatoo has pale yellow or whitish head spotting, and orange-yellow body feather barring. The lateral tail feathers are black at the base and tip, with bright orange-red and black barring on the rest of the feather, grading to yellow on the inner margin (Johnstone et al. 1998). Juvenile forest red-tailed black cockatoos are similar to adult females, except the barring on the underparts of the tail is duller. Males begin to show a reduction in head spotting and body feather barring at one to two years of age, with the first predominantly red tail feathers appearing at two and a half years. The full red tail is seen at four to five years of age (Johnstone et al. 1998). The bill of the male forest red-tailed black cockatoo darkens to black at two to three years of age. The inland red-tailed black cockatoo is very similar in appearance to the forest red-tailed black cockatoo, except that the bill of the latter is significantly longer and wider (Marchant et al. 1990) and inland red-tailed black cockatoo females have less spotting and barring (Johnstone et al. 1998). The inland red-tailed black cockatoo has a more distinct crest and sometimes has a small notch in the upper mandible, although this is not always present (Marchant et al. 1990). Ecology and behaviour As with the Baudin s cockatoo, although the general distribution of the forest redtailed black cockatoo is known, details about critical habitat areas are lacking (Chapman 2008). The cockatoo is found in the humid and subhumid south west, mainly in the hilly interior, north to Gingin (formerly to Dandaragan) and east to Mt Helena (formerly to Toodyay), Christmas Tree Well, North Bannister (formerly Wandering), Mt Saddleback (formerly to Kojonup), Rocky Gully and 17

44 the upper King River (Johnstone et al. 1998) (Figure 1.5). In recent years, the forest red-tailed black cockatoo has also been seen frequently on the Swan Coastal Plain, although the reason for this change in distribution is unknown. Figure 1.5 Distribution of the forest and inland red-tailed black cockatoo in Western Australia (Johnstone et al. 1998). Although around 90% of the forest red-tailed black cockatoo s diet is made of marri seeds (Johnstone and Kirkby 1999), it also feeds on the seeds of the jarrah (Eucalyptus marginata), blackbutt, karri, sheoak and snottygobble (Johnstone et 18

45 al. 1998). The cockatoo has also been seen stripping bark from trees to obtain invertebrate larvae (Nicholls 1905). Johnstone and Kirkby (1999) found that forest red-tailed black cockatoos feed on marri throughout the year but switch to jarrah and other foods in March and June when marri fruits are less abundant. They also appear to return to individual trees to feed on a daily basis until the supply of fruit is exhausted. While the numbers of marri and jarrah trees may be a sufficient supply for the forest red-tailed black cockatoo, nut and flower production is cyclical and therefore may limit the numbers of nesting birds in the subsequent breeding season (Johnstone and Kirkby 1999). Some seed of marri and jarrah is available all year, but the cockatoos are selective with respect to which trees they feed in and there may be differences in seed quality, nutrient value, seed size and seed fill (Johnstone and Kirkby 1999). The forest red-tailed black cockatoo nests in the hollows of marri, jarrah and karri trees and are probably monogamous for life (Marchant et al. 1990). They breed in winter and spring, and they rarely lay two eggs in one season. The incubation period lasts for days with the hen incubating and brooding while the male bird feeds with the flock and flies back to feed the hen once or twice a day (Johnstone and Kirkby 1999). The nestling period is days, after which fledglings are fed by the parents a further three to four months (Johnstone 1997). Nest hollows used by red-tailed black cockatoos are significantly wider and deeper than those used by other bird species that nest in hollows (Saunders 1982a). The inland red-tailed black cockatoo occurs in isolated populations across the centre of the continent, and in Western Australia, a population straddles the 19

46 boundary between the south-west eucalypt woodlands and the Mulga Zone. This population is expanding its range into the northern wheatbelt. Throughout the range of this species, south of the Murchison River, it feeds mainly on the seeds of the doublegee (Emex australis), but also feeds on other agricultural weeds, native vegetation including Acacia, Grevillea, Hakea and insect larvae (Saunders 1977b). The inland red-tailed black cockatoo feeds on seeding double-gees, Banksia and Grevillea spp. (Johnstone et al. 1998). The inland red-tailed black cockatoo is the only cockatoo known to breed in the autumn and spring (Saunders 1977b), with some individuals breeding at six month intervals. Productivity is very low, with slightly more than a third of nesting attempts giving rise to fledglings, and the mean weight of fledglings is only 76% of the mean weight of adult females (Saunders et al. 1985) (compared with 80-90% for most cockatoos (Cameron 2007)). This species also has only one egg per clutch. Low productivity and poor fledging weights have been associated with population declines in black cockatoos, notably in the Manmanning Carnaby s cockatoo population, which became extinct due to shortage of food (Saunders 1982b). Factors associated with the low productivity of the inland red-tailed black cockatoo may be low food quality and the agricultural treatment of doublegee and other weed species on which this species feeds (Saunders et al. 1985). 1.4 CONSERVATION STATUS, THREATENING FACTORS AND RECOVERY EFFORTS According to the 2011 IUCN Red List, both the Carnaby s cockatoo and Baudin s cockatoo are recognised internationally as endangered species and their 20

47 population trend is described as decreasing (IUCN 2011). The Commonwealth Environment Protection and Biodiversity Conservation Act 1999 lists the Carnaby s cockatoo as endangered, with the forest red-tailed black cockatoo and Baudin s cockatoo described as vulnerable (Australian Government 1999). In Western Australia, all three are specially protected under the Western Australian Wildlife Conservation Act 1950, administered by the Department of Environment and Conservation (Government of Western Australia 1950). As the forest and inland red-tailed black cockatoo are subspecies of the red-tailed black cockatoo (Calyptorhynchus banksii), they are not listed as endangered on the IUCN Red List. However, the forest red-tailed black cockatoo is listed in the Western Australian Wildlife Conservation Act 1950 as Schedule 1 Fauna (fauna that is rare or likely to become extinct). The inland red-tailed black cockatoo (C. samueli) is less endangered and is not listed under the Wildlife Conservation Act 1950, but is listed as a vulnerable species under the New South Wales Threatened Species Conservation Act (New South Wales Government 2009). There is an increasing awareness of the endangered status of black cockatoos by the public in Western Australia, although the birds are often perceived to be abundant due to their noisy nature, obvious presence and a tendency to congregate in large flocks during the non-breeding season Carnaby s cockatoo In the past 35 years the range for the Carnaby s cockatoo has reduced by a third, with local extinctions and reduced density in occupied areas (Saunders 1974a, 1990). The population is currently estimated at 20-60,000 individuals (BirdLife 21

48 International 2010b). The threats to populations of Carnaby s cockatoo include the clearing of habitat for agriculture, fragmentation of remnant habitat, soil salinity and weed invasion of remaining habitat, lack of nest hollow availability due to habitat destruction and feral bees, shooting by orchardists, competition with other species for nest hollows, poaching for illegal export and vehicle strikes (Saunders 1990; Mawson 1997; Mawson and Johnstone 1997; Saunders and Ingram 1998; Johnstone and Kirkby 2008). Despite its status as one of the world s biodiversity hotspots (Myers et al. 2000), Western Australia has undergone large-scale land development in the last 50 years at a rate that has few parallels elsewhere in the world (Johnstone and Kirkby 2005). Between 1917 and 1968, nearly 100,000 km 2 was cleared, 65% of it since 1948, when heavy machinery became available (Saunders 1982a). The remaining timber in most of the wheatbelt is along roadside verges, around buildings and in widely scattered islands of woodland interspersed by large expanses of farmland (Saunders 1982a). Agricultural clearing has had a dramatic effect on populations of Carnaby s cockatoo. In breeding areas, there remains a patchy distribution of their habitat as a result of agricultural clearing, which causes birds to spend increased amounts of time gathering food and commuting to and from their nests (Saunders 1977a). Black cockatoos are obligate tree hollow nesters (Saunders 1982a) and tree hollows in the south-west of Australia are being destroyed faster than they are being created (Saunders 1979a). A study carried out at two sites over seven years looked at differences in the breeding biology of Carnaby s cockatoos between Coomallo Creek, an area with extensively uncleared vegetation and Manmanning, an area with very little native 22

49 vegetation remaining (Saunders 1982b). The study established that the Manmanning site, with the poorer food supply and the need for cockatoos in the area to forage over greater areas to obtain sufficient food, had slower nestling weight gains than those at Coomallo Creek. Nestlings leaving the nest had lower bodyweights at Manmanning compared with those at Coomallo Creek. By 1978 the population at Manmanning had become extinct. Saunders (1977) recommends a need for corridors of native vegetation leading out from main reserves so that birds can be led to sources of food without having to expend time and energy searching. The need to search widely for food can result in a reduced breeding rate (Saunders 1977a). Not only is it important to retain linkages between remnants of native vegetation, but there is a need to re-establish corridors of native vegetation in extensively cleared agricultural areas, such as those in the wheatbelt of Western Australia (Saunders 1990). Remnant bushland in the suburbs of Perth should also be protected, as often this is the primary food source for Carnaby s cockatoos, which have adapted to the urban environment (Perry 2008). Another factor which limits the availability of tree hollows for nesting by Carnaby s cockatoo is competition with other species. Competition for hollows occurs with galahs (Cacatua roseicapilla), long-billed corellas (Cacatua tenuirostris), Port Lincoln parrots (Barnardius zonarius), regent parrots (Polytelis anthopeplus), nankeen kestrels (Falco cenchroides), barn owls (Tyto alba), boobook owls (Ninox novaeseelandiae) and feral honey bees (Apis mellifera) (Saunders 1979a). Galahs and corellas have greatly increased in number and range since the 1950s (Johnstone and Kirkby 2005). The impact of the European 23

50 feral honey bee has also increased in the past 40 years (Johnstone and Kirkby 2005). Another threat to wild populations of Carnaby s cockatoo is poaching for the pet industry. The avicultural industry in Western Australia is sometimes illegally supplied with wild-caught nestlings. As well as the direct loss of birds through poaching, damage to or loss of hollows can be a concurrent result of the poaching process. Often a hole is cut in the side of the tree, close to the floor of the hollow to obtain a nestling and cockatoos will not use the hollow again until the floor level has fallen some distance below the hole. In some situations, the tree is cut down to allow the nestling to be removed (Saunders 1979a). Two increasingly important threats to black cockatoo populations are shootings and vehicle strikes. The Carnaby s cockatoo Symposium held at the Department of Environment and Conservation in Perth in December 2008 recognised road strike and killing/harvesting as high risks to wild Carnaby s populations (Department of Environment and Conservation 2008). In the 1970s, Saunders had identified these factors as the two largest causes of death amongst tagged cockatoos (apart from predation by eagles) and warned that they would become more important as the human population increases and more clearing of native vegetation takes place (Saunders 1979b), leading to more frequent conflict with human-related activities. Of the three endemic black cockatoos in Western Australia, it is the Carnaby s cockatoo that has received the most attention with regards to conservation efforts and public profile, largely due to its habitat including the Perth metropolitan area 24

51 and therefore being visible to more people. Efforts to protect populations of Carnaby s cockatoo are orchestrated by the DEC, the Carnaby s Black-Cockatoo Recovery Team (consisting of members from DEC, World Wide Fund for Nature (WWF-Australia), the Western Australian Museum, Birdlife Australia and private landowners) and the Carnaby s Black-Cockatoo Recovery Project (operated by Birdlife Australia). Additionally, there are non-profit organisations, community groups and landholders with an interest in Carnaby s cockatoo conservation. These groups work together to achieve objectives outlined in the Carnaby s Black-Cockatoo Recovery Plan, which when written, aimed to increase population numbers of the Carnaby s cockatoo within 10 years by ensuring the species flourishes within its present range and expands into its former range (Cale 2003). Current activities undertaken by the Recovery Team and Recovery Project include nest monitoring, education and community engagement, coordination of captive breeding programs to ease pressure on illegal poaching and enforcement of laws against illegal trading and shooting. There are an increasing number of research efforts involving the Carnaby s cockatoo, as awareness of the species decline becomes more prominent. There are also mounting pressures on urban and regional developers to conserve Carnaby s cockatoo habitat (van der Moezel 2009) Baudin s cockatoo The remaining population is currently estimated to be approximately 15,000 (Johnstone and Kirkby 2008). The species has a strong association with very old ( years) marri trees which may exacerbate population declines as these 25

52 trees become less available. It is thought that this bird has disappeared from approximately 25% of its habitat, with population densities declining in another 25% of habitat area. The reporting rate for Baudin s cockatoos declined by up to 49% between and surveys (Olsen et al. 2003). The threats to populations of the Baudin s cockatoo include logging of forest habitat, feral honey bees invading nesting sites, illegal shooting by orchardists and competition for nesting sites (BirdLife International 2010a). The requirement by these birds to nest in large mature trees means that the preferential removal of such trees for timber and woodchip harvesting presents a great threat to the survival of this cockatoo species (Mawson and Long 1994). Despite its listing as a specially protected species in Western Australia, the Baudin s cockatoo is a declared Pest of Agriculture under the provisions of the Agriculture and Related Resources Protection Act 1976, administered by the Western Australian Department of Agriculture and Food. This declaration is in place due to the damage that Baudin s cockatoos inflict on commercial pome fruit (apple and pear) orchards, usually in areas where there is a local or seasonal shortage of marri seed. The declaration allows for the implementation of a management program using non-lethal methods in the Perth metropolitan region and regional Western Australian shires (Chapman 2008). Although in most years the damage caused by Baudin s cockatoo in individual orchards is minimal, localised damage can be severe and there is evidence damage is increasing in areas where marri is declining (Department of Environment and Conservation 2007). Shooting Baudin s cockatoos is illegal due to the native status of the bird; however, despite potential fines of up to $10,000 for each offence, shooting continues in some orchard areas and is considered a major threat to the species. If 26

53 illegal shooting continues at the present rate, it alone could cause the extinction of the Baudin s cockatoo (Department of Environment and Conservation 2007). The Baudin s cockatoo and forest red-tailed black cockatoo both occur in southwest Western Australia and are affected by timber harvesting, as the loss of mature forests means the loss of hollows available for nesting by these cockatoos (Saunders et al. 1985). It may be a shortage of nest sites that will limit the populations as the regenerating forest will probably provide ample sustenance (Saunders et al. 1985). Conservation efforts to protect the Baudin s cockatoo from further decline are coordinated by the Forest Black Cockatoo Recovery Team, involving members from DEC, the Western Australian Museum, BAWA, the Department of Agriculture and Food Western Australia (DAFWA) and the Western Australian Fruit Grower s Association. Current activities undertaken by the Team include collecting population data, feral honey bee eradication programs, encouraging orchardists to employ non-lethal control methods for protecting orchards from cockatoo predation and addressing illegal shooting Red-tailed black cockatoo Historically, there has been a range reduction of forest red-tailed black cockatoos of 25-30% (Mawson and Johnstone 1997) with the population in 1998 estimated at about 15,000 individuals (Abbott 1998). During the 1800s and early 1900s, forest red-tailed black cockatoos were shot for food and their beautiful tail feathers (Carter 1923), which probably had an impact on the population starting in the early 1900s (Abbott 2001). However, currently the main threats to the 27

54 species are clearing of habitat for agriculture and woodchipping (Department of the Environment and Water Resources 2007), illegal shooting and competition for nests with feral honey bees (Chapman 2008). Of the Western Australian black cockatoo species, the forest red-tail black cockatoo requires the largest nesting hollows, so it will be subject to greater pressures from forest management practices than other species that use jarrah and marri for nesting (Mawson and Long 1994). Also, they are relatively sedentary and could be extremely vulnerable to habitat loss and fragmentation (Johnstone and Kirkby 1999). Along with the other Western Australian black cockatoo species, the future of the forest red-tailed black cockatoo depends upon the development strategies undertaken in the south-west region. Habitat offsets are frequently used in an attempt to mitigate the effect of cockatoo habitat removal, however the success of offsets where there is a lag between habitat loss and the replacement of resources is likely to be low (Maron et al. 2010). Climate change is another potential threat to the forest red-tailed black cockatoo that may impact populations as a result of changes to ecosystem function (Chambers et al. 2005). Along with the Baudin s cockatoo, the recovery of the forest red-tailed black cockatoo is coordinated by the Forest Black Cockatoo Recovery Team; the activities undertaken by this team in an effort to preserve populations include eradicating feral honey bees from the south-west of Western Australia, making recommendations to federal departments regarding hotspots for developments, assessment of significant threatening agents for the populations and gathering of data on populations and habitat use (Chapman 2008). 28

55 1.5 REHABILITATION OF WILDLIFE AS A CONSERVATION STRATEGY The definition of wildlife rehabilitation can be stated as taking wild animals that are injured, sick, diseased or orphaned and providing veterinary support, with the goal of restoring them to their natural state, in the habitat from which they came (Begg and Brown 1998). Another description is the rescue of incapacitated wild animals that are considered unable to survive in the wild without human intervention and their subsequent release (Aitken 2004). It should also be noted that euthanasia is an important act of wildlife rehabilitators, when presented with wildlife that has a poor chance of survival, or when release is inappropriate from an ecological standpoint (Sleeman 2008). Wildlife rehabilitation is a large internationally growing practice (Molony et al. 2007), with great advances in wildlife medicine made over the past 20 years (Kirkwood and Sainsbury 1996). Whereas the motivation behind the rehabilitation of common species is often based on the thought that every individual animal deserves a chance to live (Finch 1998), with animal welfare often prioritised over conservation value, the objective of an endangered species rehabilitation program is to contribute to the conservation of existing wild populations. There is disagreement about the direct benefit of wildlife rehabilitation to conservation (Aitken 2004). It can be argued that the number of birds released back to the wild is too small to have an effect on wild populations; however, allowances are sometimes made for long-lived birds such as bald eagles (Haliaeetus leucocephalis) and California condors (Gymnogyps californianus), 29

56 where survivorship of adults has been found to have a major impact on population dynamics (Grier 1980). Black cockatoos are also long-lived birds, with one banded bird recorded at more than 34 years of age (Saunders and Dawson 2009). It has also been stated that if rehabilitation of birds occurs on a large enough scale, the numbers of individuals returned to the wild could have a significant impact on wild populations, particularly in cases where adult, breeding-aged birds are returned (Duke et al. 1981). Aitken (2004) makes the argument for wildlife rehabilitation as a conservation strategy, and criticises the commonly held approach to conservation, which places more emphasis on species and habitats rather than individuals. Comparisons can be made between the release of rehabilitated wildlife and the more recognised and traditional conservation strategies of reintroduction and restocking. When rehabilitated healthy wildlife is returned to the areas from where they were once found, it can be seen as a form of reintroduction which helps to maintain populations. Rehabilitation can therefore be seen as making use of what would otherwise be lost stock and actively supports the conservation rationale. Arguments against the role of rehabilitation in conservation practices involve the return of a disadvantaged individual to a wild population, thus undermining natural selection and evolutionary fitness (Aitken 2004). Critics of rehabilitation argue that treatment of compromised individuals is a form of giving the evolutionarily less fit a second chance (Kirkwood and Sainsbury 1996). However this is not necessarily so, as many animals have succumbed to injury or sickness by mischance and not as a result of being evolutionarily less fit. Also, the environment of wild animals cannot be considered to be separate from human 30

57 influences. Many studies have shown that the vast majority of wildlife admissions to rehabilitation centres are due to human activities (Fix and Barrows 1990; Deem et al. 1998; Komnenou et al. 2005; Kelly and Bland 2006). In the case of black cockatoo rehabilitation, most injuries are likely to be caused by vehicle strike and shootings. Obvious signs of underlying weaknesses or illnesses are rare and therefore this is likely to be a case of birds being unlucky rather than being physically less able to cope with natural environmental pressures. Even the strongest, fastest and healthiest of cockatoos are no match for cars and guns. Such birds would otherwise be lost to the population if it were not for efforts to treat them in captivity and return them to the wild. Kirkwood and Sainsbury (1996) concede that where harm to free-living wildlife has been caused directly or indirectly by humans, treatment and rehabilitation is an ethical course of action. The rehabilitation of injured birds also provides information on how human activities, as admission statistics may highlight issues (such as heavy traffic near feeding areas and shooting of birds in orchards) that may be affecting wild populations and maintain focus on the need for preventive measures. If the impact is found to be significant enough, then changes are warranted to prevent such injuries from occurring in the first place. Information on general morbidity and mortality in wild populations can also be gathered through rehabilitation programs (Duke et al. 1981). There have been many publications of survivorship of injured raptors that have been rehabilitated and returned to the wild, including those on bald eagles, 31

58 kookaburras (Dacelo novaeguineae) and peregrine falcons (Falco peregrinus) (Servheen and English 1976, 1979; Redig and Duke 1995; Mason 2006; Stauber et al. 2008). However, caution should be taken when comparing the post-release survivability of raptors to that of non-carnivorous birds, as there may be different fitness and condition requirements for raptors to survive after release, due to their need to actively hunt prey. Endangered species rehabilitation programs contribute to the conservation of populations not only by provision of medical treatment and preparation for return of injured wildlife to a free-living environment, but also through other elements such as research, environmental education and the provision of animals for captive breeding programs. Rehabilitation centres are often in direct contact with the public and help to raise awareness of the public to the presence of local wildlife and environmental issues (Aitken 2004). In addition to the benefits that can be gained from a rehabilitation program for endangered birds, there are also potential risks to the free-living population on release that need to be incorporated into the assessment and management of such a program. These risks may stem from species mixing in captivity and the potential for exposure to novel pathogens, other nosocomial infection, stress causing recrudescence of infection in captivity and the development of antibiotic resistance. In order to thoroughly assess these risks, a disease risk analysis of the rehabilitation program may be advocated. 32

59 1.6 REHABILITATION OF BLACK COCKATOOS AT PERTH ZOO VETERINARY DEPARTMENT The PZVD has collaborated with DEC since 2000 to provide specialised care for endangered black cockatoos requiring rehabilitation. The black cockatoo species involved in the programme include Carnaby s cockatoo, Baudin s cockatoo and the forest and inland red-tailed black cockatoo. In 2003 this collaborative arrangement was formalised, and DEC now provides financial resources to the PZVD to contribute to the management and care of these cockatoos. Each year, over 100 cockatoos are presented by DEC to the PZVD, and the numbers are increasing. Since 2003, each black cockatoo admitted to the PZVD has been given a conservation score and a prognosis score. The conservation score ranks the value of the individual bird in terms of its species and age, while the prognosis score evaluates the likelihood of success of veterinary treatment, based upon factors such as body condition, injuries and clinical signs. The scores may be updated throughout the cockatoo s diagnostic work-up and are used to provide guidelines for the triage of cases. The scoring system enables the effective allocation of resources, and is a valuable tool for evaluating the progress of each case. When the cockatoo no longer requires medication or intensive care it is sent to the Kaarakin Black Cockatoo Rehabilitation Centre, which is run by experienced carers under the supervision of DEC. Further rehabilitation is carried out in the form of increasing condition and fitness, and socialising the birds in groups before release to the wild. Cockatoos may spend any time between weeks and months in rehabilitation, depending on their injuries, the availability of a group to be 33

60 released with and the movements of wild birds. Although emphasis is placed on returning birds to the wild, in situations where the nature of the injury may preclude release, individuals may be placed in captive breeding programs (which serve to reduce the pressure on the pet black market) or within nanny flocks which help re-socialise rehabilitated cockatoos before release. The pre-release socialisation stage of the rehabilitation process results in more normal social behaviour, after the veterinary treatment period which can involve solitary housing and frequent human contact. Once cockatoos display excellent flight ability and are in good body condition they are released to the wild in groups, which is thought to increase their ability to assimilate into wild flocks. Where possible, cockatoos are returned in integrated groups to areas close to where they were found. At the release site, the cockatoos in their petpacks are placed in a line and released at once, within sight or sound of a wild flock. 1.7 CONCLUSIONS As habitat loss and fragmentation for many threatened species increase throughout the world, so do the opportunities for humans and endangered bird populations to interact, often with negative consequences for the birds (Ress and Guyer 2004). Consequently there is a growing requirement in many areas for rehabilitation facilities to treat wild birds and release them back to the wild. Although the effectiveness of rehabilitation programs in their contribution to wild populations is difficult to measure, in situations where endangered species are being directly affected by human-related activities, veterinary intervention is justified. 34

61 The black cockatoos of Western Australia face several key threats to wild populations and rely on the immediate actions of many stakeholders to secure their future. Included among the threats to these iconic birds are vehicle strike and shootings, and other causes of injury and debilitation. As part of recovery efforts, black cockatoos are treated at the PZVD and rehabilitated intensively for return to the wild. However, veterinarians currently make decisions based on experience and anecdotal evidence in order to focus efforts on cockatoos that are thought to have the greatest chance of recovery. This study focused on reviewing the 565 records from all cockatoos admitted for treatment from 2000 to 2009 to determine factors that affected the chance of recovery from injury or illness. This will improve the ability of clinicians to make informed decisions about which cockatoos are likely to have a better prognosis and direct efforts and resources towards such birds. This information may also improve welfare of cockatoo casualties, as it will provide a guide to which cockatoos have a very poor prognosis and perhaps should be euthanased on humane grounds. Other aspects of the black cockatoo rehabilitation program will also be improved upon with the results from other parts of this study, which will contribute to the further understanding of these endangered birds and the ways in which recovery efforts can be developed further to secure stable populations in the wild. 35

62 1.8 AIMS The aims of this project were: 1. To characterise the life histories of black cockatoos in rehabilitation, generally and on a seasonal basis. To follow cases presenting during the program of study, as well as to perform a retrospective study on medical records of wild black cockatoos treated at Perth Zoo since To further develop existing prognostic indicators for cockatoos admitted to the PZVD. 2. To determine the gender of birds admitted to the PZVD unequivocally, regardless of age. 3. To determine prevalence of infection with Chlamydia psittaci among wild black cockatoos undergoing rehabilitation through the PZVD. 4. To analyse the impact of signalment, specific disease conditions and injuries on the success of rehabilitation, up to the point of admission to the rehabilitation centre. 5. To establish haematologic and biochemical reference values for the three black cockatoo species endemic to Western Australia. 36

63 6. To investigate tracking options for releasable cockatoos and conduct an aviary trial to test transmitter attachments on captive cockatoos. 7. To create a genus-specific age estimation curve for black cockatoos using pentosidine analysis. 1.9 STRUCTURE OF CHAPTERS Chapter 1 presents the background to the treatment and rehabilitation of wild black cockatoos at the PZVD, describes the biology, distribution and threats to the three endemic species of West Australian black cockatoo and outlines the argument supporting the value of rehabilitation as a conservation strategy for these endangered birds. Chapter 2 presents the results of an epidemiological study of 565 wild black cockatoos admitted to the PZVD between the years 2000 and 2009, including a statistical analysis of the relationship between disease conditions, injury type, signalment, and final outcome. Chapter 2 also describes the determination of gender and likely reproductive activity of black cockatoos admitted to the PZVD between 2007 and 2009, using both surgical sexing and DNA sexing methods. Chapter 3 outlines the study to investigate the presence of Chlamydia psittaci infection among wild black cockatoos admitted to the PZVD, both on arrival at the department, after a period of hospitalisation, and during their time at the rehabilitation centre, using serological and PCR testing. 37

64 Chapter 4 presents reference values for the three species of black cockatoo endemic to Western Australia, from the results of haematologic and biochemical blood testing of 78 adult black cockatoos and describes differences between species, sex and collection method. Chapter 5 details the study exploring tracking options for black cockatoos, and outlines the aviary trial conducted on 28 captive non-release black cockatoos to determine suitable transmitter types for radio- and satellite-tracking for released rehabilitated birds. Recommendations are made on suitable transmitter attachment systems based on the results of the trial. Chapter 6 outlines the part of the study which resulted in the development of an aging curve which was constructed using pentosidine analysis of skin taken from 53 black cockatoos of known age, which is now available for use to estimate the ages of black cockatoos of unknown histories. Chapter 7 is the final chapter and consists of a general discussion of the findings of the overall study and the implications of the findings in relation to future management of black cockatoo rehabilitation. Future research is discussed, including tracking released wild black cockatoos and determining the age demographics of wild populations. 38

65 CHAPTER TWO: AN EPIDEMIOLOGICAL STUDY OF WILD BLACK COCKATOOS ADMITTED TO THE PERTH ZOO VETERINARY DEPARTMENT FROM

66 2.1 INTRODUCTION Retrospective cohort studies A retrospective or historic cohort study is a research study in which the medical records of groups of individuals are analysed for a relationship between predictor variables and fate (Mann 2003). The main benefit of conducting retrospective analyses of wildlife treatment and rehabilitation cases is that the results can be used to concentrate efforts on those animals that are more likely to recover from their injuries (Molony et al. 2007), thus optimising the use of often limited resources that are available for endangered species programs. There are also animal welfare benefits in determining the likelihood of release at an early stage of rehabilitation, as suffering will not be prolonged (Kirkwood and Best 1998) Published studies on avian rehabilitation programs Internationally, most avian rehabilitation programs involve raptor species in North America and Europe. Published retrospective studies on rehabilitated wild birds reflect this bias, with only one study published on the treatment of wild Australian raptors (Punch 2001) (Table 2.1). Few papers discuss variables associated with signalment and presenting condition in light of rehabilitation success (Ress and Guyer 2004; Molony et al. 2007) but rather list the causes and characteristics of morbidity and mortality (Fix and Barrows 1990; Sweeney et al. 1997; Deem et al. 1998; Morishita et al. 1998; Punch 2001; Wendell et al. 2002; Komnenou et al. 2005; Harris and Sleeman 2007). Trauma was the most common presenting sign in all the studies, accounting for 58-82% of cases. 40

67 Table 2.1 Retrospective studies on the rehabilitation of wild birds. Species Region Time period n % survival to release rate % trauma cases % anthropogenic causes of injury Reference Multiple raptor species California, USA (Morishita et al. 1998) Multiple raptor species Iowa, USA (Fix and Barrows 1990) Peregrine falcons Midwest USA (Sweeney et al. 1997) Multiple raptor species Florida, USA (Deem et al. 1998) Bald eagles, peregrine falcons Virginia, USA (Harris and Sleeman 2007) Multiple raptor species Western Australia (Punch 2001) Multiple raptor species Colorado, USA (Wendell et al. 2002) Multiple raptor species Greece (Komnenou et al. 2005) Multiple raptor species Multiple raptor species Blackbird, house sparrow, starling, tawny owl Tenerife, Canary Islands (Rodríguez et al. 2010) Southeastern (Ress and Guyer 2004) United States England (Molony et al. 2007) Eurasian sparrowhawks Northwestern England (Kelly and Bland 2006) 41

68 Anthropogenic effects, such as vehicle strike and bullet injury, were cited in several of the studies as the most common cause of trauma (Fix and Barrows 1990; Deem et al. 1998; Komnenou et al. 2005; Kelly and Bland 2006; Harris and Sleeman 2007; Rodríguez et al. 2010). Ress and Guyer (2004) analysed factors for their effect on outcome for raptors in the southeastern United States and found that sex, species and body mass were statistically associated with likelihood of release. Specifically, adults and birds with a lower body mass had a lower probability of release and some species were more likely to survive injury than others. The study found no significant relationship between the cause of the injury (e.g. hit by vehicle, collision/trauma, gunshot) and the outcome. The effect of the type of injury on outcome was not mentioned. In another raptor rehabilitation study, Kelly and Bland (2006) studied sparrowhawk (Accipiter nisus) rehabilitation and found that the likelihood for release was significantly lower for birds with fractures when compared with birds that had not sustained fractures. There are few examples of retrospective studies on the rehabilitation of non-raptor species. Molony et al (2007) described factors associated with survival in four English rehabilitation centres for several species, including blackbirds (Turdus merula) (n = 680), tawny owls (Strix aluco) (n = 587), starlings (Sturnus vulgaris) (n = 629) and house sparrows (Passer domesticus) (n = 543). For each of the bird species, the severity of the signs of injury or illness was found to be a significant predictor of whether an individual survived to be released or died in care. More severe injuries or illnesses were associated with lower release rates. Sex (for tawny owls only), age, time of admission, year and season of admission, body 42

69 mass on admission (for tawny owls only) and length of time in care did not have significant effects on outcome according to the final statistical models. The authors recommended that on welfare grounds, treatment should not be attempted on fractures or deep tissue wounds in passerines, and ocular damage, fractures and/or deep tissue wounds in raptors Importance of this study to the veterinary management of wild black cockatoos at PZVD The PZVD has collaborated with the DEC since 2000 to provide specialised care for debilitated wild black cockatoos that require treatment and rehabilitation. Since the beginning of the program, Perth Zoo veterinary staff have developed a high standard of care for these endangered birds. However, there is considerable expense inherent in the treatment and care of wild birds, including those costs associated with labour time, medication, surgical and diagnostic equipment. Further, the decision to hospitalise and treat a wild bird must take into account its long term likelihood of survival in the wild, balanced against the level of potentially significant stress and pain that may be experienced by the bird during the process of hospitalisation, treatment and subsequent release (Kirkwood and Sainsbury 1996). Therefore, it is important from both financial and welfare perspectives that rehabilitation programs establish potential predictive factors which are associated with outcome and examine data that may expose patterns of life histories and demographics. Results from such studies can be useful in directing resources towards those birds with a better chance of survival and reducing the number of cases with poor prognoses that are treated unnecessarily. 43

70 Although aspects of the Perth Zoo black cockatoo program have been frequently reviewed and refined since its inception, statistical analyses of past records had never been performed until this project. The results from this study will be directed towards making recommendations to the program based on statistical evidence. 2.2 AIMS The aims of this study were to: i. Identify risk factors associated with the outcome for black cockatoos undergoing treatment for rehabilitation through: a. Characterisation of life histories, including species, age and sex of black cockatoos; b. Characterisation of clinical presentations and types of injuries of admitted black cockatoos; c. Exploring temporal patterns (both annual and seasonal) of black cockatoo admissions from 2000 to ii. Map areas from which injured black cockatoos were rescued in south-west Western Australia from 2000 to 2009 and identify areas from which higher number of cockatoos were found. iii. Examine available data on those black cockatoos that have been successfully released after treatment and any available post-release data, to 44

71 determine injuries and presentations associated with eventual release and survivability. iv. Increase the understanding of morbidity and mortality in wild black cockatoos, by: a. Examination of the post-mortem findings for black cockatoos necropsied from 2000 to 2009; b. Examination of the cases of infectious disease, gastrointestinal parasitism and ectoparasite infestation in black cockatoos admitted from 2000 to MATERIALS AND METHODS PZVD black cockatoo treatment protocol All wild black cockatoos admitted to the PZVD were visually checked in their transport container for conditions requiring immediate examination or treatment, such as life-threatening haemorrhaging, open fractures or a moribund demeanour. If there was no immediate need for handling, the cockatoo was placed in a temperature-controlled, dark and quiet environment for approximately 30 minutes. All cockatoos were then given a brief physical examination under manual restraint, to check mucous membrane colour, eyes, nares, oral cavity, feathering and skin, and palpate for fractures, joint luxations or wounds. Their body condition was assessed by palpating the musculature over the keel. The cockatoo was weighed on admission and regularly thereafter, as changes in bodyweight are a good indicator of the health of the bird. During the initial examination, most 45

72 cockatoos were administered anti-inflammatories (meloxicam at 0.2mg/kg subcutaneously), antibiotics (enrofloxacin at 15mg/kg subcutaneously, diluted with an equal volume of sterile water) and warmed fluids (compound sodium lactate and 5% glucose in a 1:1 solution, given at a volume of 2-3% bodyweight subcutaneously). Anti-inflammatory medication was given to all cockatoos on admission because traumatic injuries were often difficult to detect during the brief initial examination; antibiotic medication was given to all cockatoos on admission due to the propensity for birds to develop secondary infections or septicaemia following stress or shock (S. Vitali, personal communication, 6 June 2012). Additional analgesia for fractures or other severe musculoskeletal injuries or wounds was given using butorphanol at mg/kg intramuscularly, which was repeated twice daily as required. If severe injuries were found on the initial examination, the cockatoo may have been anaesthetised briefly using isoflurane delivered by face mask for further examination, application of wound dressings or bandaging. Intravenous fluids were sometimes administered during this brief anaesthetic procedure as circulatory support for cockatoos showing signs of shock and severe dehydration. Cockatoos were euthanased during the initial examination if their injuries were incompatible with survival in the wild or captive breeding. Such injuries may have included contaminated open fractures, severe open wounds, the loss of one or both eyes or the loss of a limb. Birds that had serious injuries combined with an emaciated body condition or moribund demeanour were also considered to have a poor prognosis. Euthanasia was achieved by first anaesthetising the 46

73 cockatoo with isoflurane via a face mask and then administering intravenous pentobarbitone sodium (400mg/ml) at 2ml/kg bodyweight. If there were no life-threatening injuries, the cockatoo was housed in a hospital ward room (preferably alone, but otherwise with one or more conspecifics, each in individual cages) and was medicated orally with anti-inflammatories (meloxicam at 0.2mg/kg) and antibiotics (enrofloxacin at 15mg/kg) delivered by crop tube twice a day. After 2003, antifungal medication (terbinafine at 15mg/kg orally twice daily) was given to all birds that were handled regularly, due to some cases of aspergillosis that were seen in wild black cockatoos treated at the PZVD (see ) and the propensity for psittacines to develop aspergillosis when immunocompromised by stress or illness (Bauck 1994). The birds received supplementary feeding with granivore handrearing formula via a crop tube if inanition occurred. Other medications occasionally given include milk thistle (Silybum marianum) (5mg/kg orally twice daily) and lactulose (200mg/kg orally twice daily) for cockatoos showing signs of hepatopathy on radiographs or blood profile results. Once adequately rehydrated and stabilised (generally within 48 hours of admission), the cockatoo was anaesthetised for a thorough physical and clinical examination including blood collection and whole-body radiographs. The cockatoo was induced for anaesthesia via a rubber face mask and then intubated using a rubber 3-4mm endotracheal tube (Cook Critical Care, Bloomington, Indiana, USA). These soft rubber tubes have been found by staff at the PZVD to be appropriate for birds, given their propensity for tracheal damage by ET tubes. Isoflurane (Veterinary Companies of Australia, Kings Park, NSW) at 3-5% and 47

74 oxygen delivered at a flow rate of 1.5L/min were used to induce and maintain anaesthesia. Respiratory rate, heart rate and heart rhythm were measured with a stethoscope. Palpebral and limb withdrawal reflexes were also checked regularly. During longer procedures, a Doppler probe was used on the superficial ulnar artery to provide an audible monitor of heart rate. Supplementary heat was provided with a hot water bottle or by using a Bair-Hugger system (Augustine Medical, Eden Prairie, Minnesota, USA) and by warming the room air. During anaesthesia, the cockatoo was given a subcutaneous fluid bolus in the inguinal region of compound sodium lactate and 5% glucose (in equal volumes), at a dose rate of 2-3% bodyweight. If required, surgery was undertaken to repair fractures and soft tissue injuries. The most common method of fracture fixation was the external skeletal fixator tie-in method (Redig 2000), using intramedullary pins and fixation cross pins, which was used on long bones such as the ulna and tibiotarsus. If conservative treatment was chosen for a wing fracture, it was usually achieved by applying a figure-8 bandage. Fracture healing was evaluated by assessing radiographic changes every two weeks and by palpation for callus formation. Figure-8 bandages were left on for a maximum of 7 days and passive physiotherapy of the wing was performed at least every 5-7 days while the bandage was in place, to maintain joint range of motion. Soft tissue wounds were cleaned initially by irrigating with warmed, dilute (0.05%) aqueous chlorhexidine using an 18G needle and 20ml syringe and afterwards using sterile saline (0.9% sodium chloride solution). Unless the wound was minor, it was dressed and regular bandage changes were conducted under 48

75 general anaesthesia (GA). If gunshot pellets were present they were usually removed, although if they were difficult to access and were in a position where they were likely to remain inert, they were sometimes left in place. If necessary, lacerations were sutured using 4-0 absorbable suture material such as Vicryl polyglactin 910 (Ethicon, Somerville, New Jersey, USA). Dressings were selected depending upon the type of wound: generally, hydrogel dressings (e.g. SoloGel Wound Care Gel; Johnson and Johnson, North Ryde, Australia) were applied to dry or necrotic wounds, hydrocolloids (e.g. DuoDERM ; ConvaTec, Skillman, New Jersey, USA) were used for healthier granulating tissue and antimicrobial dressings (e.g. Iodosorb or Acticoat 7 ; Smith and Nephew, London, UK) were used for infected wounds. Secondary dressings such as Melolite (Smith and Nephew, London, UK), Fixomull (Smith and Nephew, London, UK), Coban Self-Adherent Wrap (3M, St Paul, Minneapolis, USA) and Soffban Synthetic (Smith and Nephew, London, UK) were used for extra support and protection. Biting feather lice were seen often, particularly on debilitated cockatoos. Infestations were treated with a topical pyrethrin-based spray. Ticks were seldom seen but were removed manually if present. Diagnostic blood profiles were obtained through in-house laboratory tests to determine packed cell volume (PCV), total plasma protein (TPP), estimated white cell count (WCC) and a differential white cell count. While the cockatoo was under GA for its first full examination, blood was drawn from the jugular vein (using a 23G needle and 3ml syringe) and collected into microcapillary tubes and a 1ml lithium heparin MiniCollect tube (Greiner Bio-One, Frickenhausen, 49

76 Germany) in case repeat testing was required. The blood in the microcapillary tubes was centrifuged and the PCV read directly from the spun sample. Total plasma protein (TPP) was determined using a refractometer. Blood smears were made on glass slides, air dried and then stained with the Romanowski stain Diff- Quik (Baxter Diagnostics, Deerfield, Illinois, USA) for cytological examination. In areas of the slide where there was a monolayer of red blood cells, the number of leucocytes was counted for ten fields of view, the average calculated and then multiplied by 2.5 to provide an estimated white cell count of x x Then a total of 100 leucocytes were differentiated into cell type to calculate the percentage of each type of white cell for the smear (McCracken 1993). During the time period over which this study was based, blood profiles were determined for 220 of the 565 cockatoos. The results were compared against haematologic reference values for each species (see Chapter 4). Faeces were collected from the cockatoo s cage for cytological examination and parasite screening. The faeces from 130 cockatoos from the study cohort were examined cytologically. For cytological examination, faeces were usually collected upon admission and at the end of any medications given. The faeces were smeared onto a glass slide, dried and stained with Gram stain. The slide was examined at 100X power for the presence of bacteria and the ratio of Grampositive to Gram-negative bacterium. A healthy psittacine has a predominance of Gram-positive bacteria ( 85-90%) present in the smear (Isaza 2000). From 2000 to 2009 parasitological examinations were performed on the faeces from 79 wild black cockatoos. Faeces were collected upon admission and a sample placed into a Solufix (Uropath, West Leederville, Western Australia) tube. 50

77 The faeces were macerated and left for five minutes before being filtered through a filter funnel and centrifuged at 2500rpm for five minutes. The Solufix was drained or pipetted off and approximately 5ml of saline added to resuspend the faecal pellet. The pellet was then centrifuged at 2500rpm for two minutes and the saline poured off. The last saline step was repeated and then a sample from the top of the faecal plug collected onto a microscope slide. A drop of 1% Lugol s iodine was added and mixed gently with the sample. A coverslip was placed over the sample and the slide was examined under the microscope for the presence of parasitic ova. Feather lice were also collected from wild Carnaby s cockatoos, placed in 70% ethanol and submitted to Murdoch University for identification. All rehabilitated black cockatoos that were released were first implanted with a Trovan microchip (Trovan, North Ryde, Australia) in the left pectoral muscle. The incision site was closed with Vetbond tissue adhesive (3M, St Paul, Minneapolis, USA). Prior to the commencement of this PhD project, gross necropsies were performed by the staff of the PZVD only on those cockatoos that exhibited unusual clinical signs before death. For two years of this PhD project (2008 and 2009) most cockatoos that died or were euthanased were routinely necropsied for research purposes. If lesions were found on gross necropsy, or if the cause of death appeared to be atypical (i.e. not trauma-related), tissues were fixed in 10% buffered formalin and submitted for histopathology. Fresh tissue was submitted for culture if bacterial or fungal causes were considered likely. Histopathologic examination was performed by the Animal Health Laboratory at the Department of Agriculture and Food Western Australia (DAFWA). 51

78 2.3.2 Sex determination using endoscopy and DNA methods During the earlier years of the black cockatoo program, sex was determined via external characteristics (Johnstone et al. 1998), although not always noted on the cockatoo s record. For the final two years of the dataset, after the commencement of this PhD project, sex was confirmed for 140 black cockatoos by direct visualisation of the gonads (via endoscopy or necropsy) and for 162 black cockatoos by DNA methods. Endoscopy was performed to visualise the extent of gonadal development as well as to detect ambiguities and compare efficiencies between this method and sexing by DNA. Any ambiguities between the physical appearance of the cockatoo and its sex as determined by these methods were noted. Where there were discrepancies between the results of surgical sexing and DNA sexing, the records were reviewed for possible uncertainties regarding endoscopic sexing of the individual. Given that DNA sex determination of birds is highly reliable (Jensen et al. 2003; Cerit and Avanus 2007), the DNA result was accepted as the correct result in the case of a discrepancy. Endoscopy for surgical sexing was performed on anaesthetised cockatoos, usually during their final examination prior to transfer to the Kaarakin Black Cockatoo Rehabilitation Centre. The cockatoo was positioned in right lateral recumbency and the wings extended dorsally and secured using Coban Self-Adherent Wrap. The left leg was extended forward and secured around the neck. The area for the incision (between the last rib and the pubis) was prepared using an aseptic technique. A skin incision was made and haemostatic forceps were used to make a blunt incision through the abdominal musculature into the coelomic cavity. A Stryker rigid endoscope (San Jose, California, USA) was inserted into the 52

79 coelomic cavity and the gonads visualised. The gonads were identified as either an ovary or testis (Figure 2.1) and their appearance was noted. Small, smooth testicles with an elongated shape and small, flat ovaries devoid of follicles were considered to be either mature but quiescent or immature, while large, beanshaped testes and large ovaries with follicles were considered to be mature and active. After scoping, the muscle and skin incisions were sutured using absorbable suture material, and meloxicam (at 0.2mg/kg) was given subcutaneously for analgesia. a b c d Figure 2.1 Appearance of gonads in black cockatoos at various stages of development, as seen via endoscopy (a: immature or quiescent testis (black arrow); b: immature or quiescent ovary (black arrow); c: mature active testis; d: mature active ovary). 53

80 For DNA sex determination, feathers or dried blood spots on filter paper were used. Under anaesthesia, three feathers were plucked and stored in plastic bags at room temperature until testing. Alternatively, a drop of blood was collected from the right jugular vein and placed onto Whatman s paper, which was dried and then stored at room temperature until testing. Samples were sent to Saturn Biotech, at the State Agricultural Biotechnology Centre at Murdoch University, for DNA sex determination by amplification of the CHD gene Acquisition of data The individual clinical records for all wild black cockatoos (n = 565) admitted to the PZVD during a ten-year period (January 2000-December 2009) which had been entered using the zoological software program MedARKS (ISIS, Minnesota, USA) were collated and entered onto a Microsoft Excel (Microsoft Corporation, Redmond, Washington, USA) spreadsheet (see Appendices 1a and 1b for database excerpt and key used for the database). Records were omitted from the dataset if the cockatoo had been held in captivity by a carer for more than two weeks prior to admission, or if the cockatoo had been held at the Rehabilitation Centre for education or breeding purposes prior to admission. If present on the record, the following data were entered for each case: age, sex, species, when the cockatoo was found in relation to presentation, month found, area where it was found, clinical presentation (body condition and various injuries and illnesses), unusual clinical presentations, diagnostic blood work, previous treatment received, treatment provided at the PZVD, time spent in care, the number of general anaesthetic procedures that were conducted, final outcome 54

81 (euthanased, died or survived) and findings of necropsy examination when undertaken Data management When entering data into the Excel spreadsheet, the following guidelines were used for signalment and clinical presentation: 1. The two white-tailed black cockatoo species (Carnaby s cockatoo and Baudin s cockatoo) were differentiated by the length and width of the bill (Saunders 1979c). For the purposes of this study, red-tailed species were included in one category, although two subspecies, the naso and the samueli have been encountered. 2. Prior to this PhD study, reliable aging of black cockatoos that had reached sexual maturity was not achievable, but where possible, cockatoos were grouped as juvenile or adult. The juvenile category included birds up until sexual maturity, but more commonly were those birds in their first year of life. Juveniles were identified by their physical appearance (Johnstone et al. 1998) and the presence of juvenile behaviour (e.g. begging). 3. For data entered before the commencement of this project, body condition was recorded according to the description of vets on clinical records. After 2007, a more standardised method of scoring body condition was used (Figure 2.2). Birds were grouped into one of four body condition groups (very good/overweight, fair, poor or very poor/emaciated). Body 55

82 condition rather than body mass was used when assessing the effect of variables on outcome, to account for species, age and individual differences in mass. 0 Very poor/emaciated body condition 2 Fair body condition 1 Thin/poor body condition 3 Very good to overweight body condition Figure 2.2 Body condition scoring system for black cockatoos using crosssectional size of pectoral musculature (Perth Zoo). 4. Fractures were entered if they were either grossly obvious or were diagnosed with radiography. The absence of fractures was only entered if confirmed via radiography. Fractures were categorised into different locations (bone) and type (closed/simple, closed/comminuted, open/simple, open/comminuted and old fractures). Old, healed fractures were omitted from the analysis of the effect of fracture type because it was difficult to assess their relevance to the presenting condition of admitted 56

83 cockatoos. Fracture repair was classed as either conservative (external coaptation and cage rest) or surgical. 5. Soft tissue injuries were classified as either deep or superficial. Deep soft tissue injuries included open wounds (including over fracture sites), significant bruising, swelling and major bill damage. Superficial soft tissue injuries included bruising, minor wounds and minor damage to the bill. 6. Vehicular strike was defined as trauma caused by a moving vehicle that was observed by a witness, or trauma that was thought to be highly likely to have been caused by a vehicle (i.e. the cockatoo was found on or right next to a road with injuries consistent with vehicle strike). 7. Gunshot injury was defined as injuries where bullets or bullet fragments were found associated with wounds or fractures, or where there was radiographic evidence of shotgun pellets. 8. Neurological signs included ataxia, head tilt, nystagmus and opisthotonus. Ataxia can be confused with weakness, therefore mild ataxia that was potentially due to generalised weakness was not included as a neuropathy. Paresis and paralysis were entered as separate conditions from these other neurological signs. 9. Haematologic abnormalities were determined by comparing blood results with haematologic reference values for each species (see Chapter 4). Due 57

84 to the limited sample size used to calculate the reference values, a range of values was given instead of reference intervals of two standard deviations. Therefore, the haematologic parameters from each bird were classified as abnormal if they fell outside of this range. 10. Abnormal faecal cytology was classified as a stain that showed dysbiosis (e.g. increased numbers of Gram negative bacteria), abnormally few bacteria or had organisms such as yeast that are not normally found in the faeces of healthy cockatoos. 11. Medications were included if they were administered to the cockatoo at any stage of rehabilitation. 12. Outcome was recorded as euthanased (which was further categorised into euthanased on admission or within 24 hours and euthanased following unsuccessful treatment), died (died in care despite treatment) or survived (completed all necessary treatments at the PZVD and was transferred to the Rehabilitation Centre). Some data were reclassified into similar groups to reduce the number of ambiguous categories. For example, reason for admission were grouped into 9 categories, including an undetermined category for those cases for which the reason for admission could not be ascertained from the clinical presentation, or was not recorded (Figure 2.9). Clinical signs were grouped into 12 categories, including three separate categories for different traumatic injuries because of the frequency of this type of presentation (Figure 2.10). Primary clinical signs seen in 58

85 cockatoos euthanased on the day of presentation were grouped into 14 categories, including nine groups containing different types of traumatic presentation (Figure 2.11). Fracture types were also classified into locations and listed according to frequency (Table 2.2) Statistical analysis The data were analysed using the statistical software package SPSS Statistics 17.0 (SPSS Inc., Chicago, Illinois, USA). Descriptive statistics were computed for variables such as species, age, sex, when the cockatoo was found in relation to presentation, body condition, length of time in care, treatment and outcome. The effect of different variables on the outcome of cockatoos was assessed using univariate analysis. For univariate analysis, the dependent variable needs to be dichotomous therefore outcome as the dependent variable was divided into those cockatoos that had died, whether spontaneously or by euthanasia ( died ) and those that had been transferred to the Rehabilitation Centre from the PZVD ( survived ). Initially all variables were tested separately with the dependent variable and then selected for inclusion in the multiple regression model if P The adequacy of the multiple regression model was assessed with Hosmer-Lemeshow statistics. The model is adequate if P > Odds ratios were calculated for each variable, with ratios more than one indicating that cockatoos were more likely to survive than to die or be euthanased. 59

86 Univariate analysis of risk factors against outcome was performed for the entire cohort and then for a subset of the cockatoos with fractures. Fractures were grouped for analysis of their effect on outcome according to the bone location at which they occurred, whether they were open or closed, the number of fractures present for each cockatoo and the type of repair that was given. A multivariate analysis of risk factors was performed for the entire cohort but not for the fracture subset due to the small sample size of most fracture types. Instead, a flowchart was created to display the outcomes of individual cases of black cockatoos with various fracture locations, types (open or closed), type of repair and outcome (Figure 2.13). Those cockatoos that were euthanased within the first 24 hours were omitted from the univariate and multivariate analyses because they created a bias towards a poor prognosis for recovery. Also, the criteria that are currently used in the decision-making process for immediate euthanasia is unlikely to change, meaning the usefulness of results from such analyses would be negligible. Time spent in care at the PZVD was calculated (mean, standard error and range) for the major categories of injury type for those cockatoos that survived (statistical comparisons were not performed due to small sample sizes). When calculating the time spent in care for fracture type, cockatoos were excluded if they had concurrent neurological signs or paralysis, to separate out the effect of these other conditions. Also, when assessing time spent in care for various fractures, birds were omitted if they had multiple fractures because the effect of any one of the fractures could not be isolated. Likewise, time in care was assessed for deep soft tissue injury only if there were no other major injuries (e.g. fracture) in order to 60

87 isolate the type of injury that was being analysed. Time in care was only assessed for cockatoos that survived, because this information could be useful when triaging cases according to likely costs associated with treatment Spatial distribution of cockatoo rescue locations When the suburb was recorded from which cockatoos were found and rescued, this information was collated onto urban and regional maps to show the distribution of areas from which birds were found and identify those areas which had higher numbers of birds found (Figures 2.14, 2.15). The urban map included the Swan Coastal Plain with its surrounding regions and the regional map included the entire habitat range of black cockatoo species in the south-west of Western Australia. Different coloured dots were used to represent the number of birds that were found in each area Post-release data The records of cockatoos from the study cohort that were confirmed to have been released to the wild were reviewed and their case histories summarised. Also, records were examined and case histories summarised for cockatoos that were found to have been previously treated at the PZVD; these birds were identifiable by the microchip that is implanted in every released bird (this information was also available from 2010 so this year was included for this part of the study). The records were analysed to determine what types of clinical presentations were seen in cockatoos which were eventually released, how long these birds spent in care and also to investigate the cases where birds presented to the PZVD subsequent to prior treatment, rehabilitation and release. 61

88 2.3.8 Analysis of post-mortem examination records All available records of necropsy examinations performed during the time period of the study were used to summarise the most common findings and identify the incidence of infectious disease and parasitism. Significant findings were grouped according to whether they were supported by further diagnostic testing (in addition to gross necropsy) such as histopathology, radiography or microbiology. Trauma cases were grouped according to whether the cockatoo showed internal, external or both signs of trauma. Cockatoos were described as showing evidence of internal trauma if there was haemorrhage within the coelomic cavity or organ damage that was likely to be associated with traumatic injury (Figure 2.3). External trauma was included if the cockatoo showed obvious musculoskeletal or integument damage or if fractures were diagnosed on radiograph. The most common finding after trauma was found to be mycotic infection (aspergillosis), therefore this condition was further analysed for patterns in the life histories and clinical presentations of suspect or confirmed cases. 62

89 Figure 2.3 Internal haemorrhage associated with traumatic injury in a black cockatoo (white arrows show free blood clots in the anterior coelomic cavity). 2.4 RESULTS Admission trends Signalment Carnaby s cockatoos accounted for 59% (n = 332) of total black cockatoo admissions from 2000 to 2009, while red-tailed black cockatoos accounted for 23% (n = 132) and Baudin s cockatoos accounted for 16% (n = 91) of total admissions. The species was not recorded for 2% (n = 10) of all admitted black cockatoos. For those cockatoos for which sex was recorded only by physical characteristics, there were an equal number (n = 186) of recorded males as there were females. 63

90 Where sex was confirmed by DNA and/or direct visualisation of the gonads, there were 97 (53%) males and 86 (47%) females. Twelve (6.6%) of these results (DNA and/or direct visualisation of gonads) were discrepant with the physical appearance of the bird; eight of these cockatoos had ambiguous or female physical characteristics but were confirmed as male (Figure 2.4), while four of these cockatoos appeared male but were confirmed as female. None of these discrepant birds showed mature gonad development. Figure 2.4 Immature male Baudin s cockatoo showing ambiguous physical characteristics (sex determined by DNA testing and visualisation of immature testes). 64

91 The appearance of the gonads from 116 black-cockatoos, via endoscopic sexing or visualisation of the gonads at necropsy, was recorded. Small testes were seen in 68% (n = 43) of the male cockatoos for which gonadal appearance was recorded, while 32% (n = 20) showed large mature testes. Small ovaries were seen in 57% (n = 30) and large ovaries were seen in 43% (n = 23) of the female cockatoos for which appearance was noted. More cockatoos were recorded with small testes than those with large testes for all seasonal periods; for female cockatoos that were surgically sexed, there was a higher percentage of birds with large ovaries than small ovaries in autumn and winter, compared with spring and summer when there were a higher percentage of birds with small ovaries (Figure 2.5). Month Figure 2.5 Seasonal variation in gonadal appearance among black cockatoos that were endoscopically sexed. 65

92 Age categorisation was provided for 66% of the cockatoos on the basis of physical appearance and behaviour. Juveniles made up 39% of all cockatoos admitted for which age categorisation was available, while adults accounted for 61% Temporal patterns in admissions The relative percentage of each species of black cockatoo admitted each year from (Figure 2.6) changed over the time period; notably, the number of redtailed black cockatoos seen increased between 2007 and The total number of cockatoos seen in each year increased from 18 to 151. Figure 2.6 Number of black cockatoos admitted to the PZVD from

93 Number From , the seasons during which most numbers of cockatoos presented differed according to the species (Figure 2.7). Most Carnaby s cockatoos (n = 100) presented during the summer months, with fewest birds (n = 73) presenting during the winter months. However, for both Baudin s cockatoos and red-tailed black cockatoos, the season during which most cockatoos presented was spring (n = 46 and n = 30 respectively), with fewest birds belonging to these species presenting in summer (n = 22 and n = 15 respectively) Carnaby's cockatoo Baudin's cockatoo Red-tailed black cockatoo 20 0 Summer Autumn Winter Spring Month Figure 2.7 Seasonality of the numbers of black cockatoo species admitted to the PZVD, pooled for 10 years, from In the earlier years there were fluctuations in the percentage of cockatoos admitted between seasons, however, by 2009 these differences were reduced and roughly equal percentages of cockatoos were presenting in each season (Figure 2.8). 67

94 Percentage of cockatoo admissions There was no significant difference between the years in the percentage of cockatoos that presented for each season (P = 0.78) Summer Autumn Winter Spring Year Figure 2.8 Frequency of black cockatoos admitted to the PZVD according to season from Clinical presentation The most common reason for admission of wild black cockatoos to the PZVD was trauma of undetermined origin (Figure 2.9). A substantial number of cases (n = 105, or 19%) were admitted due to an undetermined cause, with no apparent injury. Directly human-related activities (vehicle strike, gunshot and tree felling) accounted for at least 28% of cases. Additional injuries that were caused indirectly by human activity included injuries caused by collisions with inanimate objects (such as fences and posts) (n = 2) and inappropriate presentation of a chick 68

95 (n = 1). (Inappropriate presentation of a chick was defined as a healthy chick rescued and presented by a member of the public when, according to its history, it was probably already being cared for by its parents). Trauma (undetermined origin) Undetermined reason for admission Vehicle strike Gunshot Orphaned young Angular limb deformity Tree-felling (came down with tree) Predation (non-raptor) Other* Percentage of cockatoos * included metabolic bone disease (n = 1), predation by a raptor (n = 1), trauma caused by being caught in a fence (n = 1), inappropriate presentation of a chick (n = 1), trauma caused by flying into an inanimate object (n = 1), falling from a tree as a result of a storm (n = 2), and congenital deformity (n = 1). Figure 2.9 Relative frequency of reason for admission for black cockatoos treated at PZVD from The most common clinical sign seen in presented black cockatoos was deep soft tissue injury (Figure 2.10), followed by fractures and superficial soft tissue injuries. Non-traumatic conditions such as respiratory signs and diarrhoea were rare. 69

96 Deep soft tissue injury Fracture Superficial soft tissue injury Paralysis/paresis Ataxia/neurological signs Dyspnoea Diarrhoea Angular limb deformity Metabolic bone disease Regurgitation/vomiting Head tilt Upper respiratory tract signs Percentage of cockatoos presenting with sign Figure 2.10 Relative frequency of clinical signs for black cockatoos presented to the PZVD from In total, 245 black cockatoos (43.3% of admissions) presented with one or more fractures. Wing fractures were especially common; 179 (73.1%) of black cockatoos with fractures had one or more wing fractures, while 27 (11.0%) had one or more pectoral girdle fractures and 31 (12.7%) had one or more leg fractures. The frequency of fractures according to location are listed in Table

97 Table 2.2 Frequency of fractures according to location in black cockatoos presented to the PZVD from Fracture location Total number of birds with type of fracture* Percentage of total admissions (%) Wing Humerus Ulna Radius Metacarpus Digit/s Pectoral girdle Coracoid Clavicle Scapula Leg Femur Tibiotarsus Tarsometatarsus Other Pelvis Spine Keel Rib/s Skull *Some birds had multiple fractures; in these cases, all fractures have been listed in the table. 71

98 Time spent in care The amount of time spent in care at the PZVD differed according to the type of injury (Table 2.3). The mean time spent in care at the PZVD was 10 days with a range of 2-61 days. Table 2.3 Days spent in the PZVD for various clinical signs for surviving black cockatoos that presented from Condition n Mean ± SE Range Metacarpal fracture ± Humeral fracture ± Open fracture ± More than one fracture ± Combined radius/ulna fractures ± Ulna fracture ± One fracture ± Coracoid fracture ± Keel fracture ± Deep soft tissue injury ± Paralysis or paresis ± Reasons for euthanasia One hundred and fourteen cockatoos (20% of the total number) were euthanased on the day on which they were admitted. The characteristics of the injuries and 72

99 clinical presentation of birds that were selected for immediate euthanasia are summarised (Figure 2.11). The most common sign associated with a decision for immediate euthanasia was an open fracture (or fractures), which accounted for 44% of the cases. Again, non-traumatic causes for immediate euthanasia were rare. Open humeral fracture Other open fractures Multiple fractures Paralysis or paresis Moribund Severe open wounds Severe head/bill injuries Fracture not amenable to repair Severe eye injuries Congenital/developmental abnormality Open wounds and closed fractures Comminuted fractures Not recorded Crop fistula (unknown history) Percentage of cockatoos presenting with sign Figure 2.11 Primary clinical signs seen in black cockatoos euthanased on first day of presentation at the PZVD from Outcomes The outcomes for admitted cockatoos differed between years (Figure 2.12), in that more cockatoos died spontaneously in the earlier years (17% in both 2000 and 73

100 2001) compared with the more recent years (8% in both 2008 and 2009). Aside from the first year of the program (in which there was a considerably smaller sample size), the percentage of birds that were euthanased and those that survived did not change substantially over the ten-year period. Figure 2.12 Outcomes for black cockatoos treated from Univariate analysis of risk factors against outcome Some presenting signs that were rare meant that sample sizes were too small for statistical analysis. These variables included metabolic bone disease (n = 4), angular limb deformity (n = 6), dyspnoea (n = 17), upper respiratory tract signs (n = 2), regurgitation (n = 4) and diarrhoea (n = 7). 74

101 Factors that were found to be significantly associated with an increased likelihood of survival (Table 2.4) were superficial soft tissue injury (P = 0.010), more than one week in hospital (P < 0.001), more than one GA procedure (P < 0.001) and treatment with antifungal medication (P = 0.019). Factors that were significantly associated with a reduced likelihood of survival were deep soft tissue injury (P = 0.002), paralysis/paresis (P < 0.001), fracture (P = 0.002), anaemia (P = 0.026), abnormal faecal cytology (P = 0.020) and treatment with steroid medication (P = 0.009). 75

102 Table 2.4 Factors analysed against outcome for black cockatoos presented to the PZVD from Factor n Survived (%) P OR CI (95%) Baudin s cockatoo Red-tailed black cockatoo * Carnaby s cockatoo Adult Juvenile Male Female Summer * Autumn Winter Spring * Breeding season (July Dec) * Found more than 24h before * Poor body condition *

103 Table 2.4 (continued) Factors analysed against outcome for black cockatoos presented to the PZVD from Factor n Survived (%) P OR CI (95%) Been in other vet clinic Previously treated with fluids Previously treated with antibiotics Previously treated with analgesia Deep soft tissue injury * Superficial soft tissue injury * Old fracture Paralysis/paresis * Shot Fracture * Leucocytosis * Monocytosis Anaemia * Hypoproteinaemia *

104 Table 2.4 (continued) Factors analysed against outcome for black cockatoos presented to the PZVD from Factor n Survived (%) P OR CI (95%) Abnormal faecal cytology Treated with fluids Treated with antibiotics Treated with analgesia Treated with antifungals ** Treated with steroids ** Given supplementary feeding * Hospital stay of > 7 days ** More than one GA ** Presented after * Included in multivariate analysis (P < 0.25); ** Factor likely to be associated with confounding factors and not included in multivariate analysis (see below); Not included in multivariate analysis due to small sample size; Survived to transfer to the Rehabilitation Centre 78

105 2.4.3 Multivariate analysis of risk factors against outcome Following multivariate analysis, the factors that remained in the model as significantly affecting the chance of survival were superficial soft tissue injury (P = 0.008), anaemia (P = 0.004) and paralysis/paresis (P < 0.001) (Table 2.5). Hosmer and Lemeshow tests (χ 2 = 0.971, P = 0.965) indicated an adequate fit to the data for all steps of the multivariate regression model. The multivariate regression model for the data correctly predicted 108 (69.7%) out of 155 outcomes. Table 2.5 Significant variables from multivariate regression analysis of black cockatoos, excluding birds euthanased on the first day of presentation, for (survival as dependent variable). Variable P OR 95% CI Superficial soft tissue injury Anaemia Paralysis/paresis < Constant < Univariate analysis of risk factors against outcome among cockatoos with fractures After univariate analysis for all fracture categories, the only fracture category that was significantly associated with outcome was treatment by surgical repair (Table 2.6). According to odds ratios, cockatoos with fractures that were treated with 79

106 surgical repair were 2.6 times less likely to survive compared with cockatoos with fractures that did not undergo surgical repair. Table 2.6 Factors analysed against outcome for black cockatoos presenting with fractures (excluding birds euthanased on the first day). Factor n Survived (%)* P OR CI (95%) Surgical repair Open fracture Humeral fracture Ulnar fracture More than one fracture * Survived to transfer to the Rehabilitation Centre For those cockatoos with fractures, several variables were rare and therefore not included in the univariate statistical analysis. These included the fractures of certain bones (including radius (n = 30), metacarpal (n = 14), digit (n = 3), tibiotarsus (n = 11), tarsometatarsus (n = 3), femur (n = 4), keel (n = 17), coracoid (n = 17), rib (n = 2), scapula (n = 5), pelvis (n = 8), spine (n = 5) and skull (n = 5)). The sample sizes of some of these variables, such as radial fracture, were further reduced when the fracture location was isolated for analysis (i.e. there were not many cockatoos with single radial fractures). 80

107 Further, when attempting to analyse surgical repair methods, there were not enough numbers of birds with a single fracture type to determine the outcome of surgical repair based on the location of the fracture. Multiple fractures were excluded from the analysis on fracture repair so that the effect from repairing a certain bone could be examined. For this reason, and to obtain the optimum amount of knowledge from the data, the outcomes of individual cases according to their fracture type (location and closed or open) and treatment method are summarised as a flowchart (Figure 2.13). For those cockatoos that were treated for fractures, conservative treatment was chosen for 81.5% (n = 119) of cases, while orthopaedic repair was chosen for 18.5% (n = 27) of cases. 81

108 Figure 2.13 Outcomes for black cockatoos presented to the PZVD with fractures from Fractures confirmed by radiograph Multiple fractures n = 65 One fracture n = 189 Radius/ulna fracture n = 27 n = 48 n = 16 Humerus Radius Ulna Metacarpus Femur Tarsometatarsus Rib Scapula Pelvis EC n = 2 Open n = 34 n = 2 Sx n = 1 n = 1 Closed n = 10 EC n = 5 n = 3 EC n = 8 Sx EC n = * n = * Closed n = 24 n = 2 Sx n = 2 EC Open n = 4 EC n = 2 Open n = 15 n = 2 n = 7 Sx n = 2 n = 2 EC n = 17 n = 1 Closed n = 30 EC Sx n = 1 Closed n = 7 n = 2 KEY Surgical cases External coaptation (or conservative treatment) cases Number of birds that died Number of surviving birds Sx n = 1 n = 1 Digit Open n = 2 EC Open n = 1 n = 1 Sx n = 1 n = 1 Closed n = 2 n = 2 EC n = 1 n = 1 n = 1 Open n = 2 Tibiotarsus Sx Closed n = 9 n = 2 Closed n = 2 n = 2 Closed n = 8 EC n = 6 Keel Closed n = 2 EC n = 2 n = 1 n = 7 EC Closed n = 17 EC n = 1 Coracoid n = 6 Sx n = 2 Closed n = 5 EC n = 5 Clavicle Closed n = 3 EC n = 2 Open n = 14 n = 2 EC n = 4 Closed n = 5 EC n = 1 n = 3 Closed n = 12 Sx n = 1 82

109 2.4.5 Spatial distribution of location found There were apparent areas where higher numbers of cockatoos were found; these were termed hotspots. Hotspots were shown on the map as red dots, which represented an area from which more than 10 birds were found within an approximate 10km radius. On the urban map (Figure 2.14), hotspots were identified in Mundaring, Claremont, Roleystone and Byford. The regional map (Figure 2.15) did not have any obvious hotspots, although there were orange dots (representing six to nine birds rescued from the area) in the Mandurah area and north of Perth at Yanchep and Gingin. 83

110 Number of birds Figure 2.14 Map showing urban areas in Perth, Western Australia in which black cockatoos were found before presentation to the PZVD from

111 Number of birds Figure 2.15 Map showing regional areas in south-western Western Australia in which black cockatoos were found before presentation to the PZVD from

112 2.4.6 Data from cockatoos that were released The fates of cockatoos that were sent to the Rehabilitation Centre in the years were not available. However from 2007 to 2009, 60 cockatoos were released and four were kept in captivity for breeding or educational purposes. At the time of writing, 55 cockatoos at the Rehabilitation Centre were either awaiting release or their fate was undecided. Of the 60 cockatoos that were released in , 24 were found to have no apparent significant abnormalities, 19 birds showed soft tissue trauma and 15 birds presented with fractures upon initial presentation to the PZVD. One cockatoo that presented was an orphaned juvenile. Of the cockatoos with fractures, four cockatoos had a keel fracture, three had ulna fractures (with the radius intact) and two cockatoos had rib fractures. Other fracture locations for which single cockatoos presented included tibiotarsus, scapula, mandible, coracoid, clavicle and metacarpus. All of the fractures were closed. Diagnostic haematology was performed on 49 (82%) of the released cockatoos, while receiving treatment at the PZVD. Twenty of these birds showed leucocytosis, 14 had monocytosis, 21 had anaemia and six showed hypoproteinaemia. The mean time spent at the PZVD for released cockatoos was 11 days with a range of 2-48 days. The mean total time spent in care (i.e. at the PZVD and the Rehabilitation Centre) for released cockatoos for which these data were available (n = 38) was 305 days, with a range of days. 86

113 2.4.7 Post-release data Post-release survival has not yet been extensively studied for the rehabilitation program, as the only feasible method of tracking cockatoos after release is telemetry (see Chapter 5). From , only two wild black cockatoos have been admitted twice to the PZVD, identified by their microchips: 1. A Carnaby s cockatoo presented in March 2010 with haematomas following a hailstorm in Perth. Due to the unusually high number of birds hospitalised during this time because of the storm, the cockatoo was not given a full diagnostic workup but was sent to the Rehabilitation Centre within the same day. An earlier release than usual was carried out, due to the high numbers of birds that had been admitted at the same time, in otherwise good condition except for seemingly minor injuries from the hail. After its release the cockatoo was recovered one month later approximately 5 km from the release site. It was in poor body condition and radiographs showed a coracoid fracture which is likely to have been present at the first presentation. Subsequently the cockatoo did poorly in captivity, failing to gain weight and developing polyuria and was eventually euthanased. A diagnosis of pulmonary aspergillosis was confirmed with histopathology. 2. The other cockatoo that has been admitted twice was a mature female Carnaby s cockatoo, which originally presented to the PZVD in December 2008 with a fractured radius and comminuted fracture of the ulna. The fracture was treated conservatively and the cockatoo was released along with other Carnaby s cockatoos in Martin in October, Thirteen months later, the cockatoo was found dead 10 km away from the release site, in good body 87

114 condition and with ingesta within the crop. Despite a post-mortem examination and radiographs, the cause of death was not determined Post-mortem examination results and evidence of disease The findings from 161 post-mortem examinations carried out on wild black cockatoos from are summarised in Table 2.7. The most frequent postmortem finding was traumatic injury, which was diagnosed in 129 (80.1%) cockatoos. Of these cases, 85 (65.9%) showed only external signs of trauma. Three of the injured birds (2.3%) showed only internal signs of trauma and 43 (33.3%) showed evidence of both external and internal trauma. Table 2.7 Post-mortem findings for 161 black cockatoos examined at the PZVD from Significant post-mortem finding Found at gross necropsy (n) Supported by further diagnostic tests * H R M Traumatic injury: External trauma only Internal trauma only External and internal trauma Pneumonia: Mycotic Bacterial

115 Table 2.7 (continued) Post-mortem findings for 161 black cockatoos examined at the PZVD from Significant post-mortem finding Found at gross necropsy (n) Supported by further diagnostic tests * H R M Pneumonia (continued) Aspiration 1 Undetermined aetiology Other findings: Renal pathology Enteric cryptosporidiosis 1 Focal saccular oesophagitis 1 Angular limb deformity 1 1 Septicaemia 1 Polyserositis 1 Hepatopathy Tracheal/bronchial pathology 3 1 No significant findings * H = histopathology, R = radiology, M = microbiology Mycotic pneumonia (aspergillosis) Mycotic pneumonia was the most common non-traumatic post-mortem finding. Evidence of fungal pneumonia was found in 13 cockatoos (Table 2.8) and aspergillosis was usually provided as a provisional diagnosis after seeing typical 89

116 lesions (fungal mats, mould, pale yellow, green or white caseous nodules or foci, or white plaques) (Beernaert et al. 2010) on or in the lungs, airsacs or abdominal serosa at gross necropsy (Figure 2.16). Where these lesions were thought to be indistinguishable from bacterial abscesses they were categorised as unknown aetiology. Figure 2.16 Necropsy lesion associated with aspergillosis in a black cockatoo (green arrow shows white-green caseous nodule in the left lung lobe). Six mycotic pneumonia cases were supported by histopathology and one was confirmed using culture. Ten of the cockatoos had concurrent traumatic injuries. Six of the cases had been in care for only a short time (three or less days) with the rest of the cases (n = 7) developing after the cockatoo had been in care for at least 90

117 seven days. Six of the thirteen cockatoos developed aspergillosis despite treatment with antifungal medication. Most of the cases (77%) displayed clinical signs suggestive of aspergillosis (including dyspnoea, audible respiratory noises, weight loss, depression or lethargy) and six out of nine cases that had blood tests performed showed a monocytosis. Elevan of the thirteen cockatoos (85%) were in poor or very poor body condition on admission. 91

118 Table 2.8 Characteristics of mycotic pneumonia found at necropsy in 13 black cockatoos from Case ID Year Month presented Body condition on admission Clinical signs present* Concurrent trauma present Monocytosis Antifungals given Time in care (days) Supported by histopathology B February Very poor Yes No Yes No 8 - B December Poor Yes Yes - No 12 Yes B March Fair Yes Yes - No 1 - B March Unrecorded No Yes Yes No 2 Yes B December Poor Yes Yes No Yes 10 Yes B December Poor Yes Yes Yes Yes 18 Yes B March Poor Yes No Yes Yes 2 - B November Very poor Yes Yes - No 1-92

119 Table 2.8 (continued) Characteristics of mycotic pneumonia found at necropsy in 13 black cockatoos from Case ID Year Month presented Body condition on admission Clinical signs present* Concurrent trauma present Monocytosis Antifungals given Time in care (days) Supported by histopathology B April Poor No Yes No No 3 - B August Poor Yes Yes No Yes 21 Yes B February Poor Yes Yes Yes Yes 12 - B February Poor Yes Yes Yes Yes 27 Yes B March Poor No Yes - No 1 - * Included dyspnoea, audible respiratory noises, depression, lethargy, weight loss If blank, test was not undertaken. 93

120 Percentage of cockatoos with mycotic pneumoni The percentage of mycotic pneumonia cases peaked in 2003 when 7% of admissions showed evidence of infection (Figure 2.17) after which there has been a steady rate of up to two cases (1-2% of total admissions) seen per year, despite a marked increase after 2006 in the total number of black cockatoos admitted to the PZVD each year. 8 7 Change in treatment protocols after this period Year Figure 2.17 Percentage of aspergillosis cases among black cockatoos presented to the PZVD, according to year, from Parasites Gastrointestinal parasites No evidence of gastrointestinal parasites was seen in the faeces of any of 79 wild black cockatoos screened from 2000 to 2009 or during the gross necropsies of 161 wild black cockatoos during this time period. 94

121 However, ascarid nematodes (Nematoda: Ascaridia) were found in the faeces of two captive black cockatoos from the Rehabilitation Centre. One case was a wildborn forest red-tailed black cockatoo that had been kept in captivity with a rehabilitator for an unknown period, until its death following a short period of depression and recumbency. At post-mortem examination, five nematodes were found in the ascending loop of the duodenum. The cause of death was determined as clostridial necrotic enteritis and the ascarids were likely to have been an incidental finding. The second case was a juvenile captive-born, parent-reared Carnaby s cockatoo that was hospitalised with a crop fistula. During its hospitalisation, faecal screening revealed the presence of ascarid ova, although there were no associated clinical signs. Treatment with a gel containing oxfendazole led to the passing of numerous dead nematodes in the cockatoo s faeces Ectoparasites Ectoparasites were seen often on the skin of wild black cockatoos, particularly on more debilitated birds (i.e. birds that were in poor body condition and with a lethargic demeanour). The majority of ectoparasites were biting feather lice, the presence of which were mentioned in 56 records. Three lice taken from a Carnaby s cockatoo were identified as two different species: Neopsittaconirmus (Mallophaga: Philopteridae) and Franciscoloa (Mallophaga: Menoponidae). Both of these have been previously recorded in Carnaby s cockatoos (Stranger and Palma 1998). Five records also mentioned the presence of ticks, all found on the head (presumably because cockatoos were unable to self-preen this area). 95

122 2.5 DISCUSSION Admission trends The results from this study show that sex determination of black cockatoos based on physical characteristics alone may lead to an underrepresentation of males. This is a predictable result given male plumage and bill colour before reaching breeding age can be similar in appearance to females. For those birds that were sexed using DNA or direct visualisation methods, 7% of birds would have been incorrectly sexed if they had been sexed based on physical characteristics alone, although these results may also have been influenced by the experience of the clinician that examined the bird. However, where notes were made on the appearance of the gonads in these discrepant birds, all were immature and therefore these birds had possibly not yet developed adult plumage and may have been ambiguous in their physical appearance. There were a higher number of male cockatoos with small testes compared with those with large testes visualised on endoscopy, indicating a higher number of reproductively inactive males admitted to the PZVD. The data also suggest a higher proportion of females with active, mature ovaries during the autumn and winter, which may correspond with the breeding season. However, although it is likely that gonadal appearance corresponded with reproductive status, testicular size cannot be used alone to indicate reproductive activity; histopathology or cytology are required to make this distinction (Gartrell 2002). Also, data on seasonal admissions of cockatoos with differently sized gonads must be interpreted within the context of both the number of birds that were selected for 96

123 surgical sexing each season, and those for which gonadal appearance was noted, which did not include all admissions. Data concerning the estimated ages of presented cockatoos should be interpreted with caution given the subjectivity and limited sensitivity of the age determination process prior to this study. However, from the results of this study, juveniles made up a significant percentage (39%) of cockatoos admitted to the PZVD, which may be a reflection of their increased vulnerability to vehicle strike due to inexperience around roads. Chapter 6 of this thesis describes studies to provide a more accurate depiction of the age demographic of wild cockatoos entering rehabilitation using pentosidine analysis. This method can provide information not only on the juvenile/adult status of the birds, but also their actual age to an accuracy within several years (Fallon and Klandorf 2009). The temporal variation seen in the total number of black cockatoos presented to the PZVD from may be a reflection of increased rates of injuries and debilitation to wild birds but is also likely to be due, at least in part, to increased public awareness of the endangered status of the birds and the specialised program run by Perth Zoo and DEC. Since 2007, black cockatoo rehabilitators have shown a more proactive approach towards public education and this is likely to have increased the numbers of cockatoos taken into care. The number of red-tailed black cockatoos increased after 2007, when the species comprised 16% of all black cockatoo admissions, until they made up 33% of all admissions in This coincides with an increase at that time in the number of sightings of the species on the Swan Coastal Plain, which has been considered rare since the early 1900s (Department of Sustainability, Environment, Water, Population and 97

124 Communities 2011). Red-tailed black cockatoos are attracted to the berries of the introduced White Cedar (Cape Lilac) tree (Melia azederach) on the Swan Coastal Plain (Stranger 1997) but the reason for the recent dynamic expansion of their range onto the Plain is not known. Unfortunately, the growing numbers of redtailed black cockatoos in the Perth region has been seen in parallel with an increased number of casualties admitted at Perth Zoo, as they travel through the urban area and become subject to vehicle strike. The changing temporal pattern seen in outcomes of admitted black cockatoos, where the percentage of those cockatoos that spontaneously died decreased over time, is likely to be associated with the increased experience of clinicians in recognising clinical conditions that are associated with a poor prognosis at an earlier stage of the treatment process. In earlier years, the continued treatment of such conditions may have resulted in the death of cockatoos before a decision for euthanasia could be made. The variation in the numbers of Carnaby s cockatoos presenting according to season may be associated with increased movements of the birds during the non-breeding season (end of summer) towards the Swan Coastal Plain. During the non-breeding season, from January to August, there are a greater number of cockatoos present in areas of higher rainfall. This includes the Swan Coastal Plain, on which the city of Perth (and Perth Zoo) is situated, which is likely to explain the higher number of admissions during this period. Once the cockatoos reach this area, presumably both their chances of injury due to anthropogenic causes, as well as their likelihood of being found if injured, increase significantly. The seasonal distributions for Baudin s and red-tailed black cockatoo admissions were skewed towards the spring months. However 98

125 this may be confounded due to the lower numbers of these species admitted to the PZVD. The data for time spent in care may also have been affected by small sample sizes for some types of injuries; however these results may be useful for clinicians when triaging wild black cockatoos. Sample sizes were too small to examine statistically, but although the sample size for metacarpal fractures was small (n = 3), the minimum number of days a cockatoo has spent in care for this type of fracture is 19 days, which is considerably longer than the mean time of 10 days spent in care for all surviving cockatoos. This may be associated with longer healing times associated with metacarpal fractures, due to the more fragile blood vessels in this region. In common with several other wild avian rehabilitation studies (see Chapter 2.1.2) the majority of cockatoos in this study presented with trauma. The low incidence of other significant clinical and post-mortem findings suggests that most cockatoos were fit and healthy at the time of injury. These results strengthen the argument against the belief that wildlife rehabilitation leads to the reintroduction of unfit individuals to wild populations. They suggest that these birds are worth treating and rehabilitating for release, as most traumatic injuries appear to occur as a result of being in the wrong place at the wrong time rather than as a consequent of underlying poor health. A small number of cockatoos presented with injuries consistent with predation. This is likely to be due to a high rate of acute death due to this type of injury, but is also likely to be due to a low species-related susceptibility to predation in black 99

126 cockatoos. Although the wedge-tailed eagle (Aquila audax) is a natural predator of the black cockatoo (Saunders 1979b), cockatoos are probably less susceptible than other bird species to predation from mammals (such as domestic cats (Felis catus)) because of their large size and social behaviour, as these characteristics have been shown to correlate with a reduced chance of predation (Caro 2005; Møller et al. 2010). When compared to other studies, the percentage of black cockatoos presenting with fractures (43.4% of total admissions) and the distribution of fracture location (i.e. high percentage of wing fractures) was similar to those figures shown in previously published studies on avian rehabilitation in other species (Sweeney et al. 1997; Morishita et al. 1998; Komnenou et al. 2005). The higher number of wing fractures (31.7% of total admissions) compared to other fracture types (such as leg fractures, which accounted for only 5.5% of total admissions) probably reflects two factors. The first is that birds with wing fractures are unable to fly and are therefore easily caught by wildlife carers and members of the public and over-represented as a percentage of fracture types. The second factor is that the large wingspan of black cockatoos means that these bones may be more vulnerable to trauma during flight. Human activities contributed to a high number of trauma cases, and were likely to have been responsible for more, due to undocumented vehicle strikes and shootings. Few published data on mortality of wild psittacine species are available, but the high rates of human-caused morbidity and mortality in this study are consistent with the findings of studies on mortality of free-ranging raptors (Wendell et al. 2002). However, the use of these data to directly analyse 100

127 morbidity and mortality of wild cockatoo populations should be approached with caution. As has been deduced in studies of free-living raptor morbidity and mortality (Newton 1980), most deaths of cockatoos in the wild go unobserved and the deaths of those that are found and admitted for treatment are likely to be highly biased towards an association with human activities or habitation. Further, these morbidity and mortality data can only be applied to local cockatoo populations; most cockatoos in this dataset came from urban regions but the scenario is likely to differ in non-urban areas. The summary of primary signs seen in cockatoos euthanased on their first day may be useful for clinicians to review the decision making process that occurs in the first 24 hours after presentation. However, given the seriousness of the injuries that are shown in Figure 2.11, it would appear that the decisions currently being made for immediate euthanasia are well validated Analysis of factors affecting outcome The results from the univariate analysis of risk factors for outcome during the black cockatoo rehabilitation process may help to guide veterinarians in making triage decisions. Superficial soft tissue injury was associated with a greater chance of survival, which reflects the good prognosis of cases with minor injuries which are easily treated. Paralysis/paresis, fracture, deep soft tissue injury, anaemia and abnormal faecal cytology were factors that were associated with a reduced chance of survival. While the former three clinical signs are often immediately obvious, the presence of anaemia and abnormal faecal cytology may be useful additional indicators for clinicians assessing prognosis. In particular, anaemia remained a significant risk factor following multivariate analysis along 101

128 with paralysis/paresis. It must be noted however, that the multivariate model for this study was not strong in predicting death as an outcome, which indicated that there were likely to be other unidentified factors that were associated with an increased risk of death. Factors that had a P value of less than 0.25 were included in the multivariate analysis, except for the following, which were found to be associated with confounding factors: 1. Found more than one day before presentation was associated with an increased chance of survival. However, it is likely that this association was due to cockatoos with more serious injuries being presented earlier. Often cockatoos with less serious injuries were kept in care for longer periods before presentation to the PZVD. 2. More than one GA; hospital stay of more than one week and given supplementary feeding were all associated with an increased chance of survival. Rather than the repeated anaesthetic procedures, increased time in care and supplementary feeds themselves causing an increased chance of survival, this association is more likely to be due to cockatoos with treatable injuries spending longer in hospital, undergoing repeated anaesthesia for treatment and monitoring, and requiring supplementary feeding due to more intensive care. 3. Treatment with fluids; antibiotics; antifungals were associated with a greater chance of survival. However a standard treatment regime with these 102

129 combined medications was initiated in the later years for all cockatoos. During this time other factors are likely to have changed too (e.g. increased experience of clinicians treating the cockatoos and improved surgical skills). Therefore these factors may be confounding and a direct association between the treatments and outcome cannot be made for certain. An additional confounding factor may be that some birds that were euthanased early on may not have been medicated. 4. Treatment with steroids was also found to be significant against survival, however this medication was usually given to cockatoos with head trauma, which is likely to increase the chance of a poor outcome. Further, corticosteroids have been used less in birds in recent years, as avian species have been found to be particularly sensitive to the side effects of these drugs (Hess 2002). The univariate analysis of risk factors for the cockatoos diagnosed with fractures revealed that the only significant factor was surgical repair, which likely reflects the instability and severity of fractures requiring surgical repair, as well as the risks of longer anaesthetic procedures in birds. The fracture flowchart will be a useful reference for clinicians, as it shows precedents of previous treatments that have been attempted for certain types of fractures and their associated outcomes. Captivity can have a detrimental impact on stress levels and subsequently immune health (Dickens et al. 2009). This study showed that the length of time in care did not have a statistically significant effect on the outcome of rehabilitated black cockatoos, although this result may have been confounded by the number of birds 103

130 with a poorer prognosis being euthanased earlier in the treatment process. However, the prediction for post-release survival in cockatoos that have spent longer periods of time in hospital should take into consideration that cockatoos that spent longer in care may have reduced survival rates in the wild due to reduced fitness and potential habituation to predators or hazards. Although paralysis and paresis were grouped together, this type of presentation was still strongly associated with a reduced chance of survival. There were very few cockatoos that survived among the 66 birds that were unable to stand on presentation. There were only three cases that recovered after showing this initial presentation. One bird that eventually recovered was unable to stand on the first day but showed some improvement after two days, with significant improvement seen after four days. Another cockatoo that was unable to stand on the first day was standing by the following day, and a third cockatoo could not stand on the first day but was standing the following day. Some of these cases may have actually involved very weak cockatoos rather than neurologically compromised birds, however from these results it can be seen that if cockatoos are unable to stand on presentation, it is extremely rare that they will recover. Euthanasia should be considered for these birds unless improvement is seen at least within 48 hours Spatial distribution of locations at which cockatoos were found The effect of roads on local wildlife populations can be significant if not compensated for by sufficiently high birth rates (Newton et al. 1991). The spatial clusters identified in this study are consistent with areas of heavy traffic flow, such as freeways. Assuming that many of the cockatoos found in these areas had 104

131 been struck by vehicles, the hotspots may also be associated with areas in which there are higher population densities of cockatoos, as other studies have found that one of the major factors affecting road killings is good habitat conditions surrounding roads (Taylor and Goldingay 2004; Gomes et al. 2009). Certainly in the Claremont area, where one of the hotspots was identified, there are roost sites where large numbers of cockatoos aggregate during the non-breeding season (Perry 2008). As well as having high traffic flow, the roads in hotspot areas may also have feed sources attractive for cockatoos that are planted directly by the side of the road or on median strips. Another factor which may be significant is that several of the hotspots lie in the Perth foothills, which are the first heavily populated areas that cockatoos encounter along their migratory route from inland breeding grounds to the Swan Coastal Plain. A significant percentage (39%) of black cockatoos admitted to Perth Zoo are juveniles, which may not yet be wary of vehicles and may be more prone to vehicle strike when they first encounter roads. Other studies have found peaks in mortality rates of birds in periods during which juveniles are recently independent (Newton et al. 1999; Kelly and Bland 2006). These maps may provide guidance for future mitigation strategies, such as the erection of signs along hotspot roadways with the provision of a helpline number in the event that a cockatoo is struck by a vehicle. Also, the planting of cockatoo food sources in close proximity to busy roadways could be avoided, and pot holes in roads should be filled and appropriate road drainage ensured, to avoid water pools forming from which the cockatoos often drink. However, further work is required to identify the variables contributing to the clusters of cockatoo injuries at these hotspots, in order to guide appropriate preventative approaches. 105

132 2.5.4 Post-release data It is not understood why there is a low incidence of repeat admissions of black cockatoos to the PZVD. Released cockatoos may not become debilitated again despite their previous injuries and time spent in care. Alternatively, released cockatoos may have higher mortality rates but are not rescued again; the vast habitat area in which these birds are found makes this possible. The current lack of post-release monitoring means that there is no way of knowing how many individuals released from the program survive and contribute to the breeding population. The Carnaby s cockatoo that survived for 13 months in the wild after treatment for a fracture shows promise for other cockatoos that are released after such treatment, because it proves that they may survive for a significant period of time in the wild. The time period that this cockatoo survived allowed for at least one breeding period, although it would have needed to bond with a mate during this time for this to occur. Further studies involving remote tracking are essential to assess the cost effectiveness of black cockatoo rehabilitation and understand its contribution to wild, threatened populations. Although the sample size was too small for statistical comparison, it is interesting to note that despite fractures accounting for 44% of the clinical signs seen in all admitted cockatoos, only 25% of the 60 black cockatoos for which release data was available presented with fractures. Most of the released birds had either no apparent significant abnormalities on admission (40%) or soft tissue injuries only (32%). The mean time spent in care at the PZVD for these birds (11 days) was very similar to the mean time spent in care for all surviving cockatoos (10 days). The mean total time spent in care (i.e. at the PZVD and the Rehabilitation Centre) 106

133 for released cockatoos varied considerably ( days) which is probably due to the differences in time periods for which a suitable release group, release location and time becomes available for an individual cockatoo Post-mortem findings and evidence of disease The predominant post-mortem finding among 142 black cockatoos necropsied during the study period was traumatic injury (80%). Importantly, of those cockatoos affected by trauma, this was not always only externally obvious, with some cockatoos (2.3%) showing only internal evidence of trauma on post-mortem examination, and many (54%) showing both external and internal signs of trauma. This finding serves as a reminder of the extensive damage that can occur as a result of vehicle strike and other causes of traumatic injury involving large forces. Clinicians should therefore not only focus on external injuries but should also assess other parameters, such as PCV, that may indicate internal blood loss and organ damage. For example, two of the cockatoos that had only internal signs of trauma also showed both anaemia and hypoproteinaemia, which together may indicate acute haemorrhage. These findings also lend support to the current protocol of providing analgesia and fluids to all birds even if they are not showing external signs of injury. Aspergillosis is seen rarely in wild birds debilitated through injury or inanition but is the most commonly occurring disease of birds held in captivity (Redig 2008). From the results of this study, black cockatoos appear to be susceptible to contracting aspergillosis both in the wild and in captivity, as cases were seen in cockatoos upon presentation as well as those that had been hospitalised for long periods. Birds with impaired immunity, or those exposed to increased 107

134 concentrations of spores in the environment are predisposed to aspergillosis (Beernaert et al. 2010), therefore it is likely to be a secondary disease when seen in free-living birds. This is supported by the high rate of co-existing traumatic injuries in cockatoos with evidence of aspergillosis in this study and the high percentage of birds presenting in poor or very poor body condition. The total percentage of black cockatoos diagnosed with aspergillosis declined after the year This is likely due to the change in treatment protocols, where anti-fungal medication was routinely given to all black cockatoos from 2003 onwards (Figure 2.17). It is also likely that improved general supportive care contributed to the reduced number of cases seen after The majority of aspergillosis cases (92%) presented in the warmer months of November to March. Although most outbreaks recorded in wild birds have been in winter, outbreaks have also been reported in dry and dusty or hot and humid conditions, when great numbers of airborne spores can be present (Rosen 1964; Humphreys 1976). An increased risk of dehydration among debilitated cockatoos during these months may also predispose birds to further debilitation, immunosuppression and subsequent infection. Six of the eight birds that had blood taken showed monocytosis, which is a known indicator of aspergillosis (Forbes et al. 1992) and most of the cases (77%) showed clinical signs of aspergillosis, suggesting that occult infections are unusual. However, more acute forms of aspergillosis cannot be ruled out in other cockatoos necropsied during this study, as suggestive post-mortem lesions do not occur until the chronic form of the disease (Beernaert et al. 2010). Also other cases may have gone unnoticed, as not all of the cockatoos were necropsied. 108

135 In summary, clinicians should be alert to the clinical signs and haematologic indicators of aspergillosis, particularly during the warmer months of November to March. Preventative measures should be maintained, namely the reduction of stress, adequate ventilation, avoiding leaving enclosures wet after washing, using quaternary ammonium agents for disinfection, feeding fresh food and removing old food and browse material from cages (Kearns 2003). Burkholderia was cultured from an air sac swab from one cockatoo with pneumonia (Table 2.7). Although seen sporadically in poultry (Barnes 2003) this bacterium has only been reported once in psittacine birds, when B. cepacia was cultured from an African grey parrot (Psittacus erithacus erithacus) that presented with neurological signs and respiratory distress (Akkoç et al. 2008). In this study, the bacteria was cultured only to a genus level, but it represents the first report of this species from a black cockatoo. Not all cockatoos were tested for the presence of internal parasites. When some screening was performed in the earlier years of the rehabilitation program, no parasites were seen in the faeces of admitted cockatoos and therefore screening was rarely performed from 2002 until However, faecal examination of all cockatoos on admission was reinstated during 2008 and 2009 in order to gain a better understanding of parasites seen in wild black cockatoos. The absence of gastrointestinal parasites in admitted wild cockatoos is unsurprising, given that their most common parasite, the ascarid nematode, is transmitted when birds feed on contaminated flooring or ground (Doneley 2010). Ascaridosis is therefore seen commonly in captive cockatoos where birds are allowed access to ground foraging on substrate that is not changed regularly (Lloyd 2003). It follows that 109

136 ascarid nematodes were only found in the faeces of black cockatoos that had been kept in captivity, with access to dirt flooring. As a result of these findings, it is recommended that screening and treatment for gastrointestinal parastites (in particular, ascarid nematodes) is performed prior to release of rehabilitated black cockatoos to the wild, in case of parasitic burdens becoming pathogenic following the stress of release Recognition of the limitations to the study Retrospective studies by their nature often have inherent limitations; given that they are limited to cases which have already received treatment, the various groups to be studied and compared are defined by past therapeutic decisions (and in the case of this study, euthanasia decisions) (Johnston 2002). For example, the standard treatment regime of replacement fluid therapy, anti-inflammatory, antibiotic and antifungal medication has been given to all cockatoos entering the PZVD since 2005, as experience has shown that traumatic injuries and other inflammatory conditions are not always initially apparent. Dehydration is also common to wild birds entering care, due to injury, shock and stress of handling; further, the high metabolic rate of birds makes them susceptible to fluid loss after short periods of anorexia (Cousquer 2005). Because these treatments were initiated together, it was not possible to analyse their individual effect on outcome. Records showing final outcomes after rehabilitation (either release, or retention in captivity for education or breeding purposes) were not available for many surviving birds released from the PZVD. Even in the more recent years, many cockatoos that had been treated at the PZVD were still at the Rehabilitation 110

137 Centre awaiting release or completing their rehabilitation process. Therefore, cockatoos that were included in the survival category included a small number of birds that were kept in captivity as well as those destined for release. Although this is likely to have exerted a minimal effect on the data, it may have created a bias towards selection for cockatoos with injuries that would not be suitable for a wild existence, and which were consequently maintained in captivity for education and breeding. Another challenge lay in the collection of samples within the PZVD, a multistaffed institution. This was addressed by having clear protocols for sample collection in place with detailed descriptions on the methods of collection. However, the reliance upon multiple veterinarians to collect samples, differing rosters among staff and the unpredictability of admission times of wild black cockatoos led to some sample omissions. Also, the dataset was limited by the subjective and non-standardised nature of some record entries. Long periods of time were spent interpreting non-standardised entries and often important data were missing from the records. Standardising the way in which data are entered is recommended for staff working within veterinary hospitals and rehabilitation centres in case reviews or retrospective studies are carried out. These limitations and recommendations may have relevance for other studies that aim to assess the health of wild birds by taking samples for disease screening in veterinary hospitals and rehabilitation centres. It should also be noted that not all birds were radiographed or necropsied and some cockatoos may have had more serious injuries than were described in the records, such as fractures that were undetectable by palpation. Therefore the 111

138 actual number of some presenting signs may have been higher than the recorded number. 2.6 CONCLUSIONS There are growing concerns for threatened bird populations worldwide as human activities increasingly affect natural habitats and the rehabilitation of wild birds is likely to become an increasingly important practice. Although there is no common consensus about the value of rehabilitation for conservation of endangered species, there is little doubt that rehabilitation programs can at the very least greatly increase human awareness of the influence of humans on freeliving wildlife populations (Fix and Barrows 1990). Although the data produced by tagging birds and following their progress after release would be invaluable for assessing the effectiveness of rehabilitation, other indirect benefits from the rehabilitation program will remain, including its education and captive breeding potential. This study presents statistics on the morbidity and mortality of wild Western Australian black cockatoos on the Swan Coastal Plain for the first time. The number of injuries associated with human-related activities, even if underestimated, are a direct indication of the number of black cockatoos that may be affected by anthropogenic activities. From these results, it is clear that although the main threats facing wild cockatoo populations are likely to be habitat destruction and fragmentation, other human-related impacts such as vehicle strike and shootings should not be ignored, especially when measures may be taken to mitigate these effects. 112

139 CHAPTER THREE: A STUDY OF CHLAMYDIA PSITTACI INFECTION IN WILD BLACK COCKATOOS PRESENTED TO THE PZVD FOR TREATMENT AND REHABILITATION 113

140 3.1 INTRODUCTION Avian chlamydiosis is a naturally occurring, contagious and sometimes fatal disease caused by the bacterium Chlamydia psittaci. The disease in birds and humans was originally called parrot fever or psittacosis as it was believed to be contracted only from psittacine birds (Morange 1895). In 1941 the term ornithosis was used to differentiate the disease in (or contracted from) fowl from the disease in psittacine birds (Meyer 1941). The syndromes are no longer differentiated, however the disease in humans contracted from turkeys is often more severe than that contracted from psittacine birds (Andersen et al. 1997). Chlamydia psittaci is a Gram-negative intracellular bacteria and is placed in the family Chlamydiaceae of the order Chlamydiales (Harkinezhad et al. 2009). During the 1960s, the genus Chlamydia was divided into two species: C. trachomatis (Capponi and Haider 1969) and C. psittaci (Tamura et al. 1971). The species C. pneumoniae (Grayston et al. 1989) and C. pecorum (Fukushi and Hirai 1992) were added in the 1990s. In the late 1990s, taxonomic revision using DNAbased methods suggested the division of Chlamydia trachomatis and Chlamydia psittaci into separate genera, Chlamydia and Chlamydophila (Everett et al. 1999a), with the former Chlamydia psittaci placed in the genus Chlamydophila (Everett et al. 1999a; Taylor 2000). C. felis (cats), C. abortus (sheep, goats and cattle) and C. caviae (guinea pigs) were also grouped in the Chlamydophila genus. The genus Chlamydia included C. trachomatis (human), C. suis (swine) and C. muridarum (mouse and hamster) (Everett et al. 1999a). However, the decision was not widely accepted by the research community (Schachter et al. 2001) and a proposal to reunite Chlamydiaceae into a single genus Chlamydia was made a 114

141 decade later (Stephens et al. 2009). The disease in birds resulting from infection of the organism remains avian chlamydiosis. C. psittaci contains several known serovars which share an identical genusspecific antigen in their lipopolysaccharide but are distinguishable using serovarspecific monoclonal antibodies (Vanrompay et al. 1993; Andersen 1997). PCRsequence analysis and other molecular methods can also identify strains and show relationships to other known strains (Everett and Andersen 1999; Everett et al. 1999a; Laroucau et al. 2008). Additional serovars are likely to be circulating amongst wild birds as a large number of isolates have been reported from 460 avian species (Andersen 2005). The known serovars are named A, B, C, D, E, F, WC and M56 (Everett et al. 1999a) with a ninth serovar named E/B recently characterised using ompa restriction fragment length polymorphism (Geens et al. 2005a). An additional six provisional serovars have been proposed (Sachse et al. 2008). Within C. psittaci, it is apparent that certain serovars are generally found in a given group of birds (Table 3.1), however some serovars can infect more than one type of host (Vanrompay et al. 1994a; Andersen 1997; Andersen et al. 1997). For example, avian serovars are capable of infecting humans and a range of other hosts including Moorish tortoises (Testudo graeca) (Vanrompay et al. 1994a). Birds of a certain order can also be infected with strains of different serovars (Vanrompay et al. 1993). 115

142 Table 3.1 Known hosts of C. psittaci serovars (Geens et al. 2005a; Andersen and Franson 2007). Serovar A B C D E F WC M56 E/B Host/s Psittacine birds Pigeons, turkeys Ducks, swans, geese Turkeys, egrets Pigeons, ratites, turkeys Parakeets Bovine Muskrats, snowshoe hares Pigeons, ducks, turkeys Another chlamydial species, Chlamydia abortus, an abortogenic pathogen usually found in ruminants, has been isolated from a brown skua (Stercorarius antarctica) (Herrmann et al. 2000) and other avian species (Chahota et al. 2006). DNA-DNA reassociation studies have shown that C. psittaci and C. abortus are 27-85% genetically similar (Cox et al. 1988; Fukushi and Hirai 1989); indeed, C. abortus may have evolved when C. psittaci was transmitted from birds to ruminants and adapted to the new host (Pannekoek et al. 2010). C. suis and C. muridarum have also been detected in common (Falco tinnunculus) and lesser (F. naumanni) kestrels (Lemus et al. 2010). 116

143 3.1.1 Developmental cycle Chlamydiae are obligate intracellular bacteria, relying on host cells for energy needs and for the majority of nucleotide-metabolising enzymes (Andersen and Franson 2007). The life cycle is unique in that there are two functionally and morphologically distinct cell types: the elementary body and the reticulate body (Ward 1988; Oelschlaeger and Hacker 2000) (Figure 3.1). All chlamydial species undergo this biphasic cycle, differing only in the time taken to complete the cycle (Oelschlaeger and Hacker 2000). The elementary body is condensed, measuring nm in diameter, and is suited to extracellular survival. It shows no measurable metabolic activity (Oelschlaeger and Hacker 2000). The reticulate body is the larger, metabolically active cell, measuring nm in diameter, and is the replicating form which predominates most of the developmental cycle. Intermediate forms are also seen, ranging from 300 to 500nm in diameter, and are called dispersing forms or condensing forms depending on whether they are a transition from an elementary body to a reticulate body or vice versa (Andersen and Franson 2007). 117

144 Figure 3.1 Developmental cycle of chlamydiae (adapted from Vanrompay et al. 1995a). The elementary body is the infectious form, attaching to and penetrating the host cell (which is not actively phagocytic) via parasite-mediated endocytosis (Moulder 1991). Within the host cell it is held in a membrane-bound cytoplasmic vacuole derived from invagination of the host cell membrane during internalisation. If the elementary body is not then destroyed by host defence mechanisms, it may differentiate into the non-infectious reticulate body whereafter binary fission starts (Moulder 1991; Vanrompay et al. 1995a). The 118

145 environmental signals that trigger the differentiation of elementary bodies to reticulate bodies and reticulate bodies to elementary bodies are not understood (Oelschlaeger and Hacker 2000). After multiplication, chlamydiae are reorganised, through intermediate forms into new elementary bodies which may be released from the host cell (Bedson and Bland 1932). The process by which chlamydiae attach to and enter host cells remains uncertain, but it is likely that there is more than one mechanism of entry, including receptormediated endocytosis (Ward and Murray 1984). The mechanism of release from the host cell is also unclear, but chlamydiae may be released after cell lysis (Vanrompay et al. 1995a) Transmission Infected birds shed chlamydiae in faecal material, nasal secretions and eye secretions (Vanrompay et al. 1995a; Andersen 1996). The organism is environmentally labile but can remain infectious for months if protected by organic debris such as faeces (Smith et al. 2005). Transmission of C. psittaci in non-experimental situations occurs via inhalation or ingestion of contaminated material when in direct contact with infected birds, or with locations that infected birds have contaminated (Andersen et al. 1997; Andersen and Franson 2007). Inhaled infection may be acquired via nasal exudates, expired aerosol droplets, faecal aerosol droplets or dry faecal particles (Andersen and Franson 2007). Inhalation of aerosol droplets is more likely in humid, still environments that favour the persistence of droplets in the air (Burkhart and Page 1971). Experimental transmission of C. psittaci can occur by most routes including ingestion and inhalation of aerosol, and inoculation via the intramuscular, 119

146 intracerebral, intravenous, intraperitoneal and intra-airsac routes (Page 1959; Tappe et al. 1989; Vanrompay et al. 1994b; Andersen et al. 1997). However, direct transmission of aerosolised respiratory exudate is considered the primary method of transmission during outbreaks (Andersen et al. 1997). Experimental infection of turkeys via the oral route showed that ingestion is a less effective mode of transmission for C. psittaci (Page 1959). Most birds showed no response for two weeks after which clinical signs were seen, similar to those following aerosol exposure. It was thought that some birds developed infection via the intestinal route and started shedding organisms in the faeces, after which other birds inhaled the aerosolised chlamydiae. Transmission via external parasites, infection of the eye and via the egg may also be possible (Eddie et al. 1962; Burkhart and Page 1971; Page et al. 1975). Isolation of chlamydiae from eggs has been reported (Lehnert 1962; Wilt et al. 1972; Wittenbrink et al. 1993; Lublin et al. 1996), however the number of infected eggs in a flock appears low so this is probably not a very efficient method of transmission. Replication of chlamydiae within ectoparasites has not been demonstrated and it is unlikely they are a major method of transmission (Burkhart and Page 1971; Shewen 1980). Oral transmission from parent birds to their young occurs in species that regurgitate food to feed young birds (Andersen and Franson 2007). High concentrations of chlamydiae have been found in the crop and crop fluids after experimental infection in herons, egrets, turkeys and pigeons (Moore et al. 1959; Page 1959; Meyer 1965). 120

147 Many factors affect the success and rate of transmission of chlamydiae. Generally, young birds are more susceptible and morbidity and mortality are higher in this group (Meyer 1942). However, birds infected earlier in life may develop stronger immunity leading to a reduction in the severity of disease from later infections. Stress leads to immunosuppression and increased susceptibility to disease. Potential sources of stress in birds include weather changes, nesting, migration and food shortages (Burkhart and Page 1971; Smith et al. 2005). Chlamydiosis with high mortality was seen in captive violaceous euphonias (Euphonia violaceas) and hummingbirds (Amazilia amazilias) following transportation between facilities (Meteyer et al. 1992). Persistent infections are known to occur in humans, birds and sheep (Papp et al. 1998; Ward 1999; Smith et al. 2005) and are thought to occur in most species. Persistent infections may produce no or mild clinical signs and increase the period of infectivity of the host bird (Andersen and Franson 2007) Pathogenesis Experimental infection with C. psittaci has been studied in domestic poultry. When susceptible turkeys (Meleagris gallopavo) were exposed to the virulent C. psittaci serovar D by the airborne route, small numbers of organisms were found in the abdominal air sacs and mesentery within four hours after infection, with larger numbers present in the lungs and thoracic air sacs (Page 1959). By 24 hours post-infection, organisms had multiplied extensively and had reached titres of over 10 9 organisms per gram. By 48 hours after infection, low numbers of chlamydiae were present in the blood, spleen, liver, kidney and pericardial 121

148 membrane, and at 72 hours large numbers were found in the colon contents and nasal turbinates and in the bone marrow, muscle, ovaries and testes. Organisms were released into the bloodstream after multiplying in the lung and airsacs and filtered out in the spleen, liver or kidney, or were excreted via nasal and intestinal secretions (Page 1959). Turkeys experimentally infected by aerosol with serovar B (usually associated with pigeons) showed a slower development of disease than those infected with serovar D (Vanrompay et al. 1995c). Experimental infection has also been carried out using serovar A from psittacine birds and followed for pathogenesis (Tappe et al. 1989; Vanrompay et al. 1995b). Lesions seen in the air sacs and lungs were more severe in turkeys infected with serovar A and more severe pericarditis was present in turkeys infected with serovar D, however most lesions were similar with all the serovars (Tappe et al. 1989). The incubation period for, and severity and distribution of lesions caused by chlamydiosis in birds varies according to numerous factors including the exposure level, route of infection, concurrent bacterial infections, the immune status of the host, the virulence of the strain for that host species and the extent of treatment or prophylaxis (Gerlach 1994; Andersen et al. 1997; Smith et al. 2005; Harkinezhad et al. 2009). In experimentally infected young turkeys, signs of disease may be evident in 5-10 days; however in birds receiving smaller doses or in older birds, the period may be up to 2-8 weeks (Andersen et al. 1997). Differences in pathogenicity have also been found in two strains belonging to the same serovar (Vanrompay et al. 1994b). 122

149 3.1.4 Clinical disease and pathology in psittacine birds Clinical signs seen in infected birds vary widely according to the host species and the virulence of the strain. Signs in acute systemic infections with high doses of virulent strains often result in death and can include rough plumage, hypothermia, tremor, lethargy, conjunctivitis, dyspnoea, rales, coryza and sinusitis (Gerlach 1994). Emaciation, dehydration, yellow-green droppings or greyish watery droppings may also be seen. Clinical pathology findings may include an elevated white blood cell count, anaemia, elevated creatine kinase, lactate dehydrogenase and aspartate aminotransferase and elevated bile acids (Mallison 1989). Birds infected with a moderately virulent strain or with a reduced susceptibility may show lethargy, anorexia, progressive emaciation, greenish diarrhoea, conjunctivitis and high levels of urates in the droppings (Gerlach 1994; Andersen et al. 1997). Survivors can become carriers with no clinical signs (Harkinezhad et al. 2009). Immunity is generally poor and short-lived, although older birds may become more resistant to clinical disease (Andersen et al. 1997). Neurological signs including convulsions, tremors and opisthotonos have been reported in Amazon parrots and macaws (Harrison 1989). In the cockatiel (Nymphicus hollandicus), an incapacitating flaccid paresis and paralysis have been described (Gerlach 1994). A persistent keratoconjunctivitis in the absence of other clinical signs may be seen in cockatiels and budgerigars (Melopsittacus undulatus). There are no histological lesions that are pathognomonic for chlamydiosis except for the presence of intracytoplasmic inclusions known as Levinthal-Coles-Lillie 123

150 (LCL) bodies (Gerlach 1994). The severity of lesions is dependent on the virulence of the strain, the susceptibility of the host, the route of exposure and concurrent disease (McDonald and Bayer 1981). Generally lesions in systemically infected birds indicate pneumo-enteric disease. There is diffuse pulmonary congestion and there may be fibrous pleuritis. Miliary necrosis of parenchymal organs is common, probably due to the effects of chlamydial toxins (Gerlach 1994). Typical post-mortem lesions include fibrinous pericarditis, cardiomegaly, serofibrinous polyserositis, hepatomegaly, fibrinous airsacculitis, splenomegaly with necrotic foci or petechial haemorrhages (Vanrompay et al. 1995a; Andersen et al. 1997; Carter and Wise 2004). Although splenomegaly is frequently described as a common finding in avian chlamydiosis, fibrinous air sacculitis is more indicative of the disease in Psittaciformes (Gerlach 1994). Other consistent findings in psittacines include multifocal hepatic and splenic necrosis (Suwa et al. 1990). The extent of histopathological lesions depends upon the virulence of the strain, the duration of the infection, host susceptibility and the route of infection. There may be minimal histopathological changes in acute cases of chlamydiosis (Vanrompay et al. 1995a) but these can include intracytoplasmic basophilic inclusions in infected tissues (Carter and Wise 2004), splenic changes associated with histiocytosis and periportal heterophilic and mononuclear infiltrates in the liver (Vanrompay et al. 1995a). More chronic infections can lead to bile duct hyperplasia and histiocytosis, hemosiderosis and hepatic fibrosis with mononuclear infiltrates. Acute necrosis and mixed inflammatory infiltrates may be present in the kidney (Vanrompay et al. 1995a). Subacute to chronic infections 124

151 can be associated with a panmyelopathy in the bone marrow and tissue deficiencies of heterophils and macrophages, causing anaemia (Gerlach 1994). In natural settings in which the host is infected by an endemic strain, there are seldom pulmonary tissue changes, however there may be fibrinopurulent airsacculitis (Andersen and Franson 2007). Often natural infection occurs concurrently with other bacterial, viral or fungal infections which can exacerbate the severity of lesions (McDonald and Bayer 1981) Distribution and host range Avian chlamydiosis is distributed world-wide in domestic and wild birds (Andersen and Franson 2007). Experimentally, virtually any species of bird may be hosts of avian C. psittaci although some are more susceptible than others (Andersen 1997). Generally it can be assumed that all wild birds are susceptible to avian chlamydiosis (Andersen and Franson 2007). Isolates have been reported in 30 orders and 460 species (Kaleta and Taday 2003). The order Psittaciformes has the most species reported to be carrying infection, with 45% of species in the order found to be positive for Chlamydia. However the high rate of infection apparent in Psittaciformes probably reflects the large number of domestic birds belonging to this order, rather than an increased susceptibility to infection (Andersen and Franson 2007) Significance in wild populations There is little known about the effect of C. psittaci infection on wild populations, and wild birds generally show minimal clinical signs (Andersen and Franson 125

152 2007). However, inapparent or mild disease may produce signs that are not easily identified, such as decreased fertility, weak young, reduced immune function or stress causing susceptibility to other diseases (Burkhart and Page 1971; Brand 1989). It is also possible that the disease has increased significance when wild populations are under stress, due to the potential for latent infection to become reactivated, with implications for endangered wild bird populations. The potential for free-living bird populations to act as reservoirs for infection with C. psittaci may also have implications for humans, livestock and other wildlife, as crossinfection has been suggested between livestock and wild bird species (Lemus et al. 2010). Given that Chlamydia has been reported amongst so many different wild species of birds, it is likely that all wild birds are at least susceptible to infection (Andersen and Franson 2007). Occasional die-offs in free-flying populations are seen, particularly in species in which carcasses are highly visible due to their larger size and clustering as a result of flock behaviour (Andersen and Franson 2007). Free-living psittacine species exotic to Australia that have been reported with infection with C. psittaci include blue-fronted Amazon parrots (Amazona aestiva) and hyacinth macaws (Anodorhynchus hyacinthinus) (de Freitas Raso et al. 2004). The significance of avian chlamydiosis in wild Australian psittacine flocks remains unclear. In Australia, there are early reports of chlamydiosis in wild psittacine species (Burnet 1935; Beech 1953), including an incident where infected parrots (of an unnamed species) were seen to drop out of trees and die within minutes (Burnet 1939). More recent publications have focused on 126

153 outbreaks of the disease among poultry or surveys of specimens submitted to pathology laboratories (Forsyth and Beddome 1981; Reece et al. 1992) which do not give an accurate depiction of the disease status of wild populations. There are limited published studies on the occurrence of infection in wild psittacine birds in Australia (McElnea and Cross 1999) but according to the Australian Registry of Wildlife Health, C. psittaci infection was confirmed via immunohistochemistry in two sulphur-crested cockatoos (Cacatua galerita) in New South Wales after an outbreak in wild birds caused death and morbidity over a two week period (J. Hall, personal communication, 7 March 2011) Public health concerns Although direct contact with wild birds was identified as a risk factor during an outbreak of human psittacosis in New South Wales, Australia (Telfer et al. 2005), infection from pet birds (psittacines, finches, canaries and pigeons) is still considered the primary public health concern (Smith et al. 2005). Infections in domestic fowl or poultry are less of a concern, but this is contingent on the effective control of infection in production birds (Andersen and Franson 2007). Although feral pigeons in cities are often blamed for being potential sources of infection, these birds are probably not a major concern as a source of human infection (Zweifel et al. 2009). Individuals with impaired cell-mediated immunity are particularly susceptible to infection. The mortality rate in untreated patients may be as high as 20% (Carter and Wise 2004). Personnel involved with post-mortem examination of infected birds and handling of infected cultures or eggs are at particular risk (Wobeser and Brand 1982). Personal protective clothing such as long-sleeved shirts, gloves, disposable caps 127

154 and an appropriate respirator should always be worn when working with infected birds or contaminated environments. Aerosols should be kept to a minimum by spraying work areas with water or disinfectant (Andersen and Franson 2007) Diagnosis Several diagnostic methods are available for detecting avian chlamydiosis, however confirmation of the disease is notoriously difficult and can only be obtained from isolation or identification of the organism (McDonald and Bayer 1981; Fudge 1984). A presumptive diagnosis may be made on the basis of history, clinical signs and presenting pathology (Sachse et al. 2009). Clinical diagnosis to serotype level is usually difficult, as there is often insufficient antigen and commercial unavailability of the necessary serotypespecific monoclonal antibodies (Laroucau et al. 2008). The recent discovery of a new serovar and the likelihood of more serovars being present in avian populations highlights the need for improved genotyping methods (Geens et al. 2005a) such as the Multiple Loci Variable Number of Tandem Repeats (VNTR) Analysis (MLVA) method (Laroucau et al. 2008) Isolation in cell culture Isolation is the most sensitive and specific method for diagnosing avian chlamydiosis (Timms et al. 1988; Moore et al. 1991) and is widely regarded as the gold standard in chlamydial diagnosis (Thejls et al. 1994). However it is rarely performed because of cost, the requirement for labour, expertise and equipment and potential hazards to personnel (Andersen et al. 1997). Samples must be handled properly during shipment and processing to avoid loss of infectivity of 128

155 chlamydiae (Andersen and Franson 2007). When samples are submitted with an expected delay before testing, they must be maintained at 4 C in a suitable transport medium, such as sucrose/phosphate/glutamate (SPG) medium supplemented with foetal bovine serum, antibiotics and a fungal inhibitor (Spencer and Johnson 1983). Culture can be performed in either embryonated chicken eggs or tissue culture and avian strains are relatively easy to isolate (Andersen et al. 1997). Tissues (air sacs, spleen, pericardium, heart, liver and kidney) are collected aseptically at necropsy, or oropharyngeal and cloacal swabs are taken from live birds. Faeces are not suitable for isolation attempts (Andersen et al. 1997). Because chlamydiae may be shed intermittently, a single negative result from a swab is not a reliable indication that the bird is free of infection (Vanrompay et al. 1995a). Antimicrobial therapy administered two to three weeks before testing may return a false negative result (Grimes et al. 1987) Polymerase chain reaction (PCR) Several PCR techniques have been developed for the detection of chlamydiae in tissues and swabs (Hewinson et al. 1997; Messmer et al. 1997; Robertson et al. 2009). Some studies have shown that PCR is more sensitive at the detection of Chlamydia than cell culture, however the highly sensitive nature of PCR can be problematic unless the correct precautions are taken (McElnea and Cross 1999). Most PCR primers have been developed for ribosomal DNA which has multiple copies, giving the test increased sensitivity (Andersen and Franson 2007). Nested multiplex PCR has been developed to increase sensitivity in detection of 129

156 chlamydiae (Messmer et al. 1997) but often such procedures require increased manipulation in the laboratory which can magnify the risk of cross-contamination. Another multiplex PCR has been developed with increased specificity because it requires a match on two DNA segments (Everett et al. 1999b). Real-time PCR methods have been described for C. psittaci (Geens et al. 2005b; Menard et al. 2006; Pantchev et al. 2009) and in contrast to conventional PCR methods, enable the quantification of DNA, rather than producing a positive or negative result Serology The results of serological tests will vary because of the length of time the bird has been infected, the concentration of circulating antibodies and host characteristics (Fudge 1997). Interpretation of results can be difficult because antibodies from previous exposure may last for several months (Geens et al. 2005b) and previous treatment with antibiotics may interfere with antibody formation, producing false negatives. Serological tests are less valuable as a diagnostic tool for wild birds, however they are useful in epidemiological studies for determining the prevalence of infections (Andersen and Franson 2007). The most commonly used serological test in birds is the complement fixation (CF) test, which detects immunoglobulin G (IgG) antibody to the lipopolysaccharide (LPS) antigen (Andersen and Franson 2007). The usefulness of CF tests in birds is species dependent (Grimes 1989). A confirmed diagnosis of chlamydiosis usually involves a four-fold increase in antibody titres with a CF test. This is usually impractical with wild birds, therefore a presumptive diagnosis may be made when high antibody titres and clinical signs are seen in a flock (Grimes and Arizmendi 1996). 130

157 There are also several commercially available enzyme-linked immunosorbent assay (ELISA) kits for detecting IgG antibodies in avian species, which are rapid and sensitive. These utilise genus-specific LPS antigen (Prukner-Radovcic et al. 2005) and include the ImmunoComb Avian Chlamydia psittaci Antibody Test Kit (Biogal-Galed Labs, Kibbutz Galed, Israel) (see Chapter ), Chlamydiapsittaci-AK-EIA (Roehm Pharma, Darmstadt, Germany) and RIDASCREEN C. psittaci (Biopharm, Darmstadt, Germany). Other serological methods include latex agglutination and elementary body agglutination, which were developed for use with psittacine bird sera and detect immunoglobulin M (IgM) and IgG antibody (Andersen et al. 1997). Methods that measure IgM, which is produced during acute infection, may only require a single positive result for a presumptive diagnosis (Grimes and Arizmendi 1996) and are useful for detecting current or recent infections (Andersen et al. 1997) Immunoassays for antigen detection Most commercially available immunoassay kits have been developed and are used primarily for the detection of Chlamydia trachomatis in humans, however because of the utilisation of the family-specific LPS antigen, they are suitable for the detection of C. psittaci (Sachse et al. 2009). These antigen detection kits include Clearview Chlamydia MF (Unipath Ltd, Bedford, United Kingdom), IMAGEN (Celltech, Berkshire, United Kingdom), Chlamydiazyme (Abbott Diagnostics, Chicago, USA) and IDEIA (Dako Diagnostics Ltd, Cambridgeshire, United Kingdom). 131

158 The advantages of using immunoassays to detect antigen over cell culture include the shorter time required, lower cost and the ability to test in the absence of viable organism (Sachse et al. 2009). However the tests have been shown to lack sensitivity and specificity when tested on birds (Vanrompay et al. 1994c) and often other diagnostic methods such as serology or PCR are used in preference to immunoassays. For example, when using the Clearview test, both false negatives and false positives have been described compared with tissue culture (Vanrompay 2000) Immunohistochemistry and histochemical staining Smears can be made from faeces, cloacal swabs, conjunctival scrapings, choanal or nasal swabs and impression smears from liver, spleen, pericardium, kidney and lung. Prepared smears can be stained with one of several staining procedures including modified Machiavello, modified Gimenez, Giemsa or modified Ziehl- Neelsen (MZN) (Andersen 2004). Detection of stained chlamydial bodies is dependent upon experience and samples with low levels of infection may be overlooked (Sachse et al. 2009). Immunofluorescent antibody testing (IFAT) can also be used to detect antigen on smears, by targeting anti-lps or anti-momp (major outer membrane protein) antibodies, although if numbers are too low for easy detection, immunoassays or PCR-based methods are more likely to be successful (Sachse et al. 2009). Chlamydiae can be demonstrated in histological specimens using a variety of staining procedures including Giemsa (Sachse et al. 2009). The technique is sensitive but the interpretation requires experience because other bacteria may also be stained (Andersen et al. 1997). Immunohistochemical (IHC) staining 132

159 utilises monoclonal antibodies against LPS or MOMP and is more sensitive than histochemical staining. IHC staining can detect chlamydial antigen in liver, spleen, lung, intestine, air sac, adrenal gland, bone marrow, conjunctiva and capillary endothelium of many tissues and organs. It can confirm that inclusion bodies seen in the cytoplasm of macrophages, hepatocytes, capillary endothelia and bile duct epithelia are chlamydial (Andersen and Franson 2007). However, cross-reactions with some bacteria and fungi may occur (Andersen and Franson 2007) DNA microarray technology DNA microarrays, which consist of gene sequences attached to solid surfaces and exploit DNA hybridisation to determine gene expression, are a recently developed diagnostic tool capable of detecting more subtle intra-species variations than standard PCR (Sachse et al. 2009). The technology has been used to develop an assay for the detection and differentiation of Chlamydia spp. (Sachse et al. 2005) and more specifically, for the development of a genotyping assay for C. psittaci which has identified previously untyped strains (Sachse et al. 2008). Microarray technology is expensive and not readily available for commercial diagnostic use and there is an increased risk of contamination due to only one reaction tube being used (Robertson et al. 2009) Treatment and control The treatment of avian chlamydiosis can be difficult (Smith et al. 2005). Individual birds undergoing treatment should be isolated for the course of antibiotic therapy, which is usually 45 days, although recent studies have shown elimination of infection in experimentally inoculated cockatiels after a 21 day 133

160 treatment period (Andersen and Franson 2007; Guzman et al. 2010). Doxycycline, a tetracycline antibiotic, is the treatment of choice for birds with avian chlamydiosis (Smith et al. 2005) and may be given by several different routes. Doxycycline is better absorbed and has a longer half life than other tetracyclines and therefore may be given in lower doses or administered less frequently (Smith et al. 2005). The drug is available in various injectable formulations, some that are suitable for administration by the intramuscular route (Gylsdorff 1987) and all of which can be irritant at the injection site. The recommended dose rate is mg/kg by intramuscular injection every five to seven days for the first four weeks and then every five days for the duration of treatment (Gylsdorff 1987). Doxycyline can also be administered orally using monohydrate or calcium-syrup formulations. Dose rates vary between species but a dose rate of 25-30mg/kg PO every 24 hours is recommended for cockatoos and macaws (Smith et al. 2005). An intravenous form of doxycycline is available in the United States (Andersen and Franson 2007). Treated birds should be monitored for signs of immunosuppression and inhibition of autochthonous flora (Doneley 2010) and other signs of doxycycline toxicosis, which include depression, anorexia, green-stained urine and elevated hepatic enzymes in the serum (Smith et al. 2005). During outbreaks affecting large numbers of birds, treatment of individual birds may not be not feasible. According to pharmacologic studies, doxycycline hyclate in water at doses of mg/L for cockatiels, mg/L for Goffin s cockatoos and 800mg/L for African gray parrots will maintain therapeutic concentrations (Powers et al. 2000; Flammer et al. 2001). However, in-water medication requires that birds are drinking a minimum volume of water, 134

161 which may be unpredictable with some desert species or when alternative water sources are available. Also, in-water medication can result in variable blood concentrations in individual birds which may mean that some birds receive subtherapeutic levels while others may receive high levels of medication (Flammer et al. 2001). A recipe for medicated seed for budgerigars has also been described (Flammer et al. 2003). Oxytetracycline has been used to treat avian chlamydiosis, with dose rates of 75mg/kg subcutaneously every three days recommended for Goffin s cockatoos, Amazon parrots and blue and gold macaws (Flammer et al. 1990). Repeated injections are not advisable because the drug causes irritation at the injection site, therefore injectable oxytetracycline is recommended only for initiating treatment in unwell or anorexic birds. Chlortetracycline may also be used to medicate feed such as mash diets, pellets and fruit (Arnstein et al. 1968; Arnstein et al. 1969; Landgraf et al. 1982; Flammer et al. 1989). Intake of chlortetracycline-medicated feed should be monitored carefully because it is unpalatable (Andersen and Franson 2007). An important part of treatment for avian chlamydiosis is supportive care for the patient, which may include a heated environment, fluid therapy and supplementary nutrition. Lactulose can be given in cases of hepatitis (Harrison 1989). Treatment of wild birds may be indicated for individuals of rare species, groups in captivity or during quarantine or translocation periods (Andersen and Franson 2007). For control of infection in free-flying birds, reduction in the amount of 135

162 infective material may be achieved though the collection and incineration of carcasses, however the risk of spreading the infection to new locations should be considered (Andersen and Franson 2007). Depending on the country or region involved, official notification of outbreaks may be required and control strategies may be directed by government authorities. In Australia, lists of notifiable diseases vary between states. Control of the disease in captive flocks is dependent on quarantine measures, hygiene and treatment of outbreaks. Routine quarantine protocols should encompass isolation of newly acquired birds for a minimum of 45 days, in an area with separate air space from other birds. Prophylactic treatment with doxycycline is sometimes recommended for prevention of chlamydiosis during the quarantine period, however this may be associated with the generation of resistant strains of C. psittaci and other bacteria (Smith et al. 2005). Cages should be cleaned with disinfectants that are effective against C. psittaci which include 1:1000 dilution of quaternary compounds, 70% isopropyl alcohol, or a 1:32 to 1:100 dilution of household bleach (sodium hypochlorite) (Smith et al. 2005). Disinfectants can cause respiratory irritation in birds and should be used in a well-ventilated area (Smith et al. 2005). A vaccine against C. psittaci is not commercially available. Producing a vaccine effective against chlamydiosis is difficult because of the different serovars and strains (Vanrompay et al. 1995a) and because important host defence mechanisms against the organisms are not completely understood (Vanrompay et al. 1999). A MOMP-based DNA vaccine against a C. psittaci serovar A strain has been demonstrated in turkeys, however further research into protective host immune 136

163 mechanisms is needed for improvement of such DNA vaccines (Vanrompay et al. 1999). 3.2 AIMS The aims of this part of the project were: i. To determine the prevalence of C. psittaci infection in black cockatoos admitted to PZVD, upon admission, after one week of hospitalisation and during rehabilitation at the Black cockatoo Rehabilitation Centre. ii. To assess the risk to black cockatoos of contracting C. psittaci infection during hospitalisation and rehabilitation. iii. To make recommendations for the management of black cockatoos during hospitalisation and rehabilitation to reduce the risk of contracting C. psittaci infection. 3.3 MATERIALS AND METHODS Thirty-one wild black cockatoos were sampled for serological testing upon admission and during the second week of hospitalisation (see Table 3.2). Twentythree of these birds were then sampled while being held at the Black cockatoo Rehabilitation Centre before their release to the wild. For PCR testing, 23 of these same birds were sampled upon admission, 19 were sampled after one week 137

164 in hospital and 23 at the Rehabilitation Centre. Where possible, the second sample was taken at least seven days following the first sample, which allowed enough time for either the cockatoo to start producing antibodies if infection was picked up in the hospital, or for the cockatoo to start shedding antigen if a latent infection was activated by the stress of handling and captivity. Cases were selected that were predicted to have a good chance of reaching rehabilitation, in order to follow cockatoos through the rehabilitation process and collect the full set of samples. Six cockatoos that had been resident at the Black Cockatoo Rehabilitation Centre for at least 12 months were sampled once for serology and PCR. These cockatoos originated in the wild but were not releasable because of permanent disabilities caused by traumatic injury. 138

165 Table 3.2 Number of cockatoos tested for C. psittaci infection during various stages of rehabilitation. Stage of rehabilitation Test Carnaby s cockatoo (n) Red-tailed black cockatoo (n) Baudin s cockatoo (n) Admission to PZVD Serology PCR Second sample post-admission Rehabilitation Centre (pre-release) Resident at Rehabilitation Centre Serology PCR Serology PCR Serology PCR The first sample collection was performed under anaesthesia during the initial physical examination, which usually occurred within 24 hours of admission. The second sample collection was taken during the second anaesthesia, where possible at least seven days following the first sample collection. The third sample was taken while the cockatoo was at the Rehabilitation Centre, either when the bird came in for a recheck at the PZVD, or at the Rehabilitation Centre under manual restraint. 139

166 3.3.1 Serological testing ImmunoComb Avian C. psittaci Antibody Test Kit Serological testing for the presence of IgG antibodies was performed at the PZVD, using the ImmunoComb Avian C. psittaci Antibody Test Kit. The kit is based on solid phase dot-elisa technology. The kit consists of a comb-shaped card containing 12 teeth, pre-punched discs of filter paper and a multi-chamber developing plate containing 12 rows of six wells (labelled A-F). Antigen is present on a solid phase on each tooth of the comb-shaped card. Blood on discs of filter paper is mixed with diluent in the first row of wells in the multi-chamber developing plate. The test spots on the Comb are then incubated with the sample in the well and any IgG antibodies that are present in the sample bind to antigens on the test spots. The Comb is then transferred to the next well, where any unbound antibodies are washed from the test spot. The Comb is moved to the next well to allow a reaction with an anti-avian IgG alkaline phosphates conjugate, which binds to antigen-antibody complexes at the test spot. The Comb is washed two more times and then moved to the last well where a colour develops due to an enzymatic reaction. The intensity of the colour depends on the antibody level in the test sample. The colour change can be read directly as a colour (ranging from white to dark grey) or by using the calibrated CombScale included in the kit, which results in a Comb score of 0-6 (6 indicating a high positive result). 140

167 Sample collection Blood was collected from cockatoos which were anaesthetised or restrained manually, depending on other procedures being undertaken at the same time. If the cockatoo was anaesthetised for sample collection, blood was collected from the jugular vein using a 23G needle and 3ml syringe, after preparation of the area using 70% ethanol. If sample collection was performed under manual restraint, blood was collected from the medial metatarsal vein using a 23G needle and 3ml syringe, or a 1ml tuberculin syringe. Following blood collection, blood was placed immediately onto both sides of the pre-punched discs of ImmunoComb filter paper. After application of the blood spot, the ImmunoComb paper was dried at room temperature, placed in a small plastic bag and refrigerated at 4 C for up to 6 months until testing Sample testing using the ImmunoComb Kit The Immunocomb testing was carried out as per the manufacturer s instructions PCR testing Sample collection Cockatoos were either anaesthetised or held under manual restraint for sampling the conjunctiva, choanae and cloaca. If sampling was performed under manual restraint, a metal mouth gag wrapped in adhesive plaster was used to hold the bill open to swab the choanae. Two swabs were taken from each cockatoo. When swabbing for Chlamydia, the second swab is more reliable as the first swab removes any exudate that may be present on the surface of the mucous membranes, leaving the second swab to sample the columnar epithelial cells for 141

168 which chlamydial bacteria have a tropism (Wyrick and Raulston 2007). For this reason, the first swab was stored while the second swab was sent for testing. A plastic handled sterile cotton-tipped swab was rolled over the mucosa of the conjunctiva (see Figure 3.2), choanae and cloaca (respectively) and then stored in the sterile plastic container provided with the swab at 4 C. Sampling was repeated using a wooden-handled sterile cotton-tipped swab, which was immediately inserted into an Eppendorf tube (Eppendorf South Pacific Pty Ltd, North Ryde, Australia) containing 5μL β mercaptoethanol (Sigma-Aldrich, Castle Hill, Australia) and 500μL Buffer AL lysis buffer (Qiagen, Doncaster, Australia). The tube was sealed and refrigerated at 4 C for up to 17 months until testing. Detection of DNA is possible at least 18 months following sample collection (T. Robertson, personal communication, 8 January 2010). The samples were packed on ice and sent to the School of Veterinary Science, Melbourne University for testing. 142

169 Figure 3.2 Swabbing the conjunctiva of a Carnaby s cockatoo for C. psittaci PCR testing DNA extraction, PCR and DNA sequencing DNA extraction, PCR testing, high resolution melt (HRM) curve analysis and DNA sequencing were conducted at the Asia Pacific Centre for Animal Health at The University of Melbourne, using methods described by Robertson et al. (2009). DNA extraction was performed by adding 15µL of QiaexII matrix (Qiagen, Doncaster, Australia) and 300µL of 70% ethanol were added to the swabs in suspension. The matrix was resuspended by vortexing then loaded into a multispin MSK100 column (Axygen, Union City, California, USA), centrifuged and washed once with 600µL buffer RW1 (Qiagen, Doncaster, Australia) and twice with 500µL buffer RPE (Qiagen, Doncaster, Australia). The columns were dried by centrifugation then overlaid with 50µL of distilled water. The columns were incubated and centrifuged again. 143

170 PCR testing was conducted using the 16SG oligonucleotide set (Messmer et al. 1997) in a RotorGene thermocycler 6000 (Corbett Research, Mortlake, NSW, Australia). The reactions were subjected to initial denaturing of 2 min at 95ºC, and then to 35 cycles of 20s at 96ºC, 20s at 58ºC (annealing temperature) and 20s at 72ºC, followed by an extension period of 2min at 72ºC. All samples were tested in triplicate. Distilled water and yolk stalk from an egg inoculated with Chlamydiaceae transport media were used as negative controls in each reaction. HRM curve analysis was performed using the software Rotor-gene (Qiagen, Doncaster, Australia). Amplicons were purified using the QIAquick PCR purification kit (Qiagen, Doncaster, Australia), eluted in 30µL of water and subjected to automated sequencing (Big Dye Chemistry, Applied Biosciences, New York, USA) in both directions using the oligonucleotide set 16SG (5µmol/L) as for PCR. The nucleotide sequences were compared using the Australian National Genomic Information Service (ANGIS) ( Statistical analysis Statistical analyses were performed using EpiTools (Sergeant 2012). Group sensitivity was calculated to determine the probability of detection of disease using the two different tests (serology and PCR). Sample size requirements to detect various prevalences of the disease in the population using the two tests were also calculated. 144

171 3.4 RESULTS There were three seropositive results recorded among the 75 serum samples tested (Table 3.3). The positive results came from three cockatoos, housed permanently at the Black cockatoo Rehabilitation Centre. The three birds were housed in the same aviary. One of the birds, a Baudin s cockatoo, had been at the Rehabilitation Centre for more than 24 months and had a CombScale result of 2 (positive). The second positive result was from a red-tailed black cockatoo which had been at the Rehabilitation Centre for more than 24 months and showed a CombScale result of 1 (positive). Another Baudin s cockatoo that had been resident at the Rehabilitation Centre for at least 4 years also showed a CombScale result of 1 (positive). Table 3.3 Results of C. psittaci diagnostic testing using serology for antibody detection and PCR for antigen detection for black cockatoos during different stages of rehabilitation. Stage of rehabilitation Seroprevalence PCR positive Admission to PZVD 0/34 0/23 Second sample post-admission 0/23 0/23 Rehabilitation Centre (pre-release) 0/17 0/17 Resident at Rehabilitation Centre 3/6 0/6 Total 3/80 0/69 145

172 As a result of PCR testing, Chlamydia DNA was detected in six swabs. Four of these contained insufficient DNA for sequencing. The remaining two swabs were purified for sequencing and speciation, however the initial sample quality of one of these swabs was not optimum, which led to incomplete binding of primers during the PCR. The other swab showed a match with the Chlamydiacae family, but not with C. psittaci itself. According to a calculated group sensitivity, there was an 81% probability of detecting seropositivity in the 34 birds on admission if it were present at a prevalence of 5% or more, assuming a test sensitivity of 95% (Bendheim et al. 1998). For the same prevalence, there was a 65% probability of detecting C. psittaci using PCR on the 23 birds tested on admission, assuming a test sensitivity of 90%. At the same estimated prevalence, to achieve a 95% confidence of freedom level, a sample size of 61 animals would be required using serology (assuming 95% test sensitivity) and 65 animals would be required using PCR (assuming 90% sensitivity). 3.5 DISCUSSION Results of testing The sample size of cockatoos in this study reduces the ability to draw definitive conclusions regarding wild populations, however, given the absence of positive results from wild black cockatoos admitted to the PZVD during the study, the prevalence of C. psittaci infection amongst wild black cockatoos is probably low. A larger sample size would be required to estimate the true seroprevalence (n = 418 if the true seroprevalance was 5%) or prevalence of the infection (n =

173 using PCR if the prevalence was 5%), however sampling and testing these numbers would have been cost-prohibitive. It should also be noted that treatment with antibiotics may cause false negatives to C. psittaci testing (Smith et al. 2005). All cockatoos are administered antibiotics at least once (and usually for at least several days) after their admission to the PZVD, which may have affected the second round of PCR sampling. Further, it was not always possible to perform the second sample collection after at least seven days, as sometimes cockatoos were anaesthetised sooner than this. This may have resulted in some cockatoos being tested before they were able to mount an antibody response leading to false negatives on serological testing. The positive antibody results seen in three resident cockatoos at the Black cockatoo Rehabilitation Centre could possibly represent false positive results, as antibody titres may vary between species (Bendheim et al. 1998). However, these cockatoos may have been potential carriers of C. psittaci infection and therefore resident birds should be considered as a risk of infection for wild black cockatoos during their stay at the Rehabilitation Centre. It seems likely that the positive results seen in six of the swabs after PCR testing were false positive results. PCR testing is notoriously sensitive and false positives commonly occur as a result of contamination (McElnea and Cross 1999). Furthermore, all of the cockatoos that showed positive PCR results appeared healthy and were seronegative; although this does not exclude the possibility of Chlamydophila infection, it supports the likelihood of false positive PCR results. A true positive result can produce a trace positive PCR, either as a result of very 147

174 low initial numbers of Chlamydia cells or due to sample storage affecting the quality of the DNA present (H. Blacker, personal communication, 21 July 2010). It seems improbable that there was poor template DNA due to incorrect sample handling, as swabs were always immediately placed in buffer solution and refrigerated Historical cases of chlamydiosis among wild black cockatoos admitted to the PZVD There have been no confirmed cases of chlamydiosis among wild black cockatoos admitted to the PZVD, although some suspect cases have been identified via necropsy: 1. In February 2001, a Carnaby s cockatoo was presented to the PZVD with depression, lethargy and green faeces around the vent. The cockatoo was in poor body condition and was anaemic with heterophilia. Gross necropsy revealed splenomegaly and a congested liver with multiple granulomatous reactions throughout the parenchyma. Histopathology was not performed but chlamydiosis was a differential diagnosis. 2. In May 2009, a juvenile female Carnaby s cockatoo presented depressed and unable to perch. The cockatoo was in poor body condition and had retained crop contents. ImmunoComb testing for antibodies to C. psittaci and PCR testing were negative. The cockatoo was found dead three days after admission. Gross necropsy showed hepatomegaly with pale discolouration and a mottling to the colour of the spleen. There was a focal brown discolouration in one of the lung lobes. IHC staining of liver and spleen 148

175 impression smears showed strong positive intracellular staining consistent with chlamydial infection, however there was similar staining on the negative control, indicating a possible cross reaction with a different organism. Histology revealed acute hepatic necrosis and acute tubular necrosis. IFAT was negative on impression smears of the liver and spleen. The liver histopathology was consistent with a very acute chlamydial infection but the case could not be confirmed by IHC or IFAT. 3. In June 2010, a Carnaby s cockatoo presented with a chronic wing fracture and was euthanased due to its injuries and poor body condition. Gross necropsy revealed hepatomegaly and splenomegaly. Immunohistochemistry showed light but diffuse distributions of antigen throughout the liver and spleen. Immunofluorescent antibody testing was negative from impression smears taken from the liver and spleen. From these results, it was surmised that the cockatoo was carrying Chlamydia organisms but it was uncertain if the infection was causing disease. C. psittaci infection was not confirmed in any of the above cases and given that these were the only three suspect cases out of more than 565 black cockatoos that presented to the PZVD during this time period, it seems unlikely that C. psittaci infection is prevalent in local wild populations. However, for most of this time period, only those cockatoos with suspicious or atypical clinical signs were selected for necropsy and further testing, and no antemortem testing was conducted routinely. This selection bias may have caused missed diagnoses of other cockatoos that were carrying subclinical infections. 149

176 3.5.3 Outbreak of chlamydiosis at the Kaarakin Black Cockatoo Rehabilitation Centre The risk of cockatoos contracting chlamydiosis during their stay at the Rehabilitation Centre was highlighted after an outbreak of the disease among cockatoos housed at the Centre. In June 2010, a male Carnaby s cockatoo that had been at the Rehabilitation Centre was found dead in its aviary, after appearing fluffed and lethargic earlier in the day. The cockatoo had recently been transferred from a different aviary due to aggressive behaviour displayed by cage mates. Gross necropsy showed splenomegaly and hepatomegaly. Histopathology revealed necrotising splenitis with intranuclear chlamydial elementary bodies and moderate diffuse necrotising hepatitis. Immunofluorescent antigen testing from impression smears of the liver and spleen were positive. The cockatoo had been admitted to the PZVD with a coracoid fracture and sent to the Rehabilitation Centre four months previously, during which time both Immunocomb and PCR testing (taken from a combined conjunctival/choanal/cloacal swab) were negative. Fourteen days after the death of this cockatoo, two other Carnaby s cockatoos from the same aviary presented with signs consistent with chlamydiosis. The first bird was an immature female cockatoo which presented with severe weight loss, bright green faeces and lethargy. The cockatoo returned positive ImmunoComb antibody, Clearview antigen and PCR test results and was euthanased due to its very poor body condition. Necropsy revealed diffuse hepatitis and splenitis (Figure 3.3) with intralesional chlamydial organisms. Immunofluorescent antigen testing from impression smears of the lung, liver and spleen was positive. 150

177 Figure 3.3 Carnaby s cockatoo (Perth Zoo record number B ) diagnosed with chlamydiosis showing splenomegaly (white arrow) and hepatomegaly (green arrow). The second cockatoo from the same aviary, a young female cockatoo, showed severe weight loss, bright green faeces and lethargy. This bird had a positive ImmunoComb result and a negative Clearview antigen test result. Clearview testing was performed on a combined conjunctival/choanal/cloacal swab. PCR testing also showed a negative result. The cockatoo was isolated and treated with doxycycline (at 100mg/kg IM), enrofloxacin, meloxicam, milk thistle and lactulose. A second Clearview test taken one week later was also negative, although this may have been a false negative caused by antibiotic treatment. Doxycycline injections were repeated weekly for seven weeks, when clinical 151

178 signs were no longer present and body weight had increased. The cockatoo made a recovery to apparently full health. A third Carnaby s cockatoo from the same aviary did not show clinical signs of chlamydiosis but tested positive to ImmunoComb antibody testing. The cockatoo was isolated and given weekly intramuscular doxycycline injections (at 100mg/kg) for seven weeks, after which both ImmunoComb and Clearview testing were negative. Quarantine restrictions were imposed for 45 days on neighbouring aviaries, those in close proximity to the infected aviary and the aviary from which the index case cockatoo had been recently transferred. The infected aviary was left empty for this period, the top layer of substrate removed and the aviary and its furniture thoroughly disinfected. Utensils and feeding bowls from this aviary were discarded. Personnel entering and leaving the quarantine aviaries sprayed footwear with appropriate disinfectant and washed their hands with antibacterial hand wash. Pathways in front of the quarantine aviaries were taped off and signs were placed to warn of restricted movements. Movements of birds in and out of the facility were suspended for the quarantine period. Birds in quarantined aviaries were placed on in-water doxycycline treatment (at 0.4mg/ml) for 45 days. Two weeks following the end of the treatment course, testing was carried out on 22 birds from the quarantine aviaries and from an aviary containing cockatoos about to be released to the wild. The results from ImmunoComb and Clearview testing for all of these samples were negative. PCR testing was also negative. 152

179 The origin of the infection of these birds remains unknown, but possible sources include wild birds (e.g. galahs, Eolophus roseicapilla) that frequently alight on and around the wire-topped aviaries. If such birds were carriers of C. psittaci then inhabitants of the aviaries could contract infection via airborne secretions or aerosolised dried faecal matter. Other sources of infection include other black cockatoos or other psittacine species that were housed at the Rehabilitation Centre, that may have been carrying latent infections. Alternatively, the index case may have had a chronic infection that it obtained in the wild, which was activated after stress induced by cage mate aggression and an enclosure change. However, this seems less likely given the cockatoo s seronegative status when tested upon admission to the PZVD Management recommendations The outbreak of C. psittaci infection and the positive serological responses of two resident black cockatoos at the Kaarakin Black Cockatoo Rehabilitation Centre highlighted the risk of infectious disease spreading among this captive collection of endangered psittacine birds and served as a reminder of the implications if birds carrying infection were released. This is particularly important given results of testing of admitted wild birds through the program suggested a low prevalence among local wild populations. It is recommended that programs designed for the release or translocation of captive birds into the wild should incorporate quarantine, diagnosis of mortalities and intermittent testing of live birds to help prevent the release of birds affected by chlamydiosis (Andersen and Franson 2007). Outbreaks of chlamydiosis 153

180 amongst large collections of captive psittacines can have disastrous consequences (Ecco et al. 2009). Several management actions are recommended for management of infectious disease risk in the post-hospitalisation phase of the black cockatoo rehabilitation program: 1. Movements out of a quarantine aviary in which new admissions are housed should be restricted until new birds have been housed for a sufficient time period for clinical signs of disease to develop (45 days). The quarantine aviary should be serviced by dedicated personnel who will clean boots before and after entry with an appropriate bactericidal disinfectant, such as TriGene II (MediChem International, Kent, UK). Hygiene protocols should include washing of hands with antibacterial hand wash before and after servicing aviaries. 2. Personnel at the Rehabilitation Centre should be appropriately educated regarding quarantine practices for the prevention of infectious diseases. The use of personal protective equipment is to be reinforced and workers should be aware of suggestive clinical signs of chlamydiosis for early detection of potential cases and prevention of zoonotic transmission. Any workers coming from other avian rehabilitation centres or from private homes containing pet birds should change clothing and wash their hands thoroughly before entering the premises. 154

181 3. Any dead birds from the Rehabilitation Centre should be refrigerated and transported to the PZVD immediately for post-mortem examination. Cockatoos with suspicious signs consistent with chlamydiosis should be necropsied by Perth Zoo vets taking the necessary precautions (e.g. wearing particulate masks, wetting down feathers to reduce airborne particles) to mitigate the risk of contracting zoonotic infection. 4. There should be restrictions on the housing of species other than Calyptorhynchus at the Rehabilitation Centre. The host specificity of C. psittaci means that a particular strain may show disease in black cockatoos but be inapparent in another species (Andersen and Franson 2007). At the time of the outbreak, two blue-and-gold macaws (Ara ararauna) were housed at the facility, within mesh contact of the infected aviary row. These birds tested negative to PCR testing for C. psittaci but are a reminder of the potential for other species to harbour subclinical infection. 5. Where possible, aviaries should be modified to reduce the possibility of close proximity and disease incursion from wild birds. For example, structural modifications should be incorporated to deter wild birds from alighting on top of black cockatoo aviaries. 6. Serological and antigen testing of rehabilitated black cockatoos should be carried out prior to release to the wild for exposure to C. psittaci, to prevent possible spread of infection from the Rehabilitation Centre to wild, naïve populations. The in-house kits Immunocomb and Clearview are appropriate for such screening. If identified, seropositive birds or birds 155

182 that test positive to Clearview should be tested using PCR to determine their shedding status. PCR-positive birds should be isolated and treated appropriately, then retested following treatment to determine whether infection has been cleared. 3.6 CONCLUSIONS The negligible incidence of positive serological and PCR tests over the course of this study, coupled with the absence of confirmed cases of infection among wild cockatoos rehabilitated over the last 10 years, suggest that the risk of wild black cockatoos contracting C. psittaci infection during hospitalisation at the PZVD is very low. However, black cockatoos awaiting release or in permanent residency at the Rehabilitation Centre may be at some risk of contracting the infection due to the presence of potential carrier birds at the Rehabilitation Centre identified through serological testing, the proximity to wild birds that may act as reservoirs for infection and various stressors that are inevitable when housing wild psittacine birds in captivity. The release of rehabilitated wild animals can pose significant health risks to wild populations (Tribe and Brown 2000), however most releases are still undertaken with a lack of awareness of the disease risks (Vogelnest 1998). It is vital that pertinent health issues are taken into account when rehabilitating and releasing wild animals, particularly those with an endangered status. The results of this study will help to guide management decisions associated with the captive housing of wild black cockatoos during rehabilitation for release to the wild. 156

183 Future captive breeding programs for the release of black cockatoos to the wild will also benefit from these findings. 157

184 CHAPTER FOUR: HAEMATOLOGIC AND PLASMA BIOCHEMICAL REFERENCE VALUES FOR THREE SPECIES OF BLACK COCKATOOS (CALYPTORYNCHUS SPP.) This chapter is modified from the following paper: Le Souëf A, Holyoake C, Vitali S, Warren K Hematologic and serum biochemical reference values for three species of Western Australian black cockatoos (Calyptorhynchus spp.). Journal of Avian Medicine and Surgery. 27(1):

185 4.1 INTRODUCTION The assessment of haematologic and biochemical parameters are an important diagnostic tool for avian patients (Samour and Howlett 2008), as clinical signs of ill-health in birds are often subtle (Hochleithner 1994). However, the interpretation of clinical pathology results can be difficult without species-specific reference values. Reference intervals are still not available for many avian species (Low et al. 2006) and are currently scarce for black cockatoos (Polo et al. 1998; ISIS 2002a, 2002b). The establishment of reference values for Western Australian black cockatoos will improve the standards of veterinary care for these endangered birds, both in the treatment of injured wild cockatoos and in the health monitoring of birds in captive breeding programs. Such baseline clinical pathology data may also be useful in future health surveillance studies of wild populations. 4.2 MATERIALS AND METHODS The Carnaby s cockatoos (n = 34), Baudin s cockatoos (n = 22) and forest redtailed black cockatoos (n = 22) used for the study were from either the Perth Zoo captive collection or were housed at a rehabilitation centre in Western Australia (Table 4.1). Cockatoos from the rehabilitation centre had been captive for at least 12 months but were originally wild birds from various regions in the south-west of Western Australia. All birds were clinically normal, had no recent history of ill-health, and were at least 12 months old. All birds were on a diet of mixed seeds, vegetables, and native browse. 159

186 Table 4.1 Origins of the black cockatoos sampled for compilation of haematologic and biochemical reference ranges. Origin Carnaby s Baudin s Forest red-tailed cockatoo cockatoo black cockatoo Zoo Rehabilitation centre Total Carnaby s cockatoos and Baudin s cockatoos were sexed according to distinct dimorphic characteristics which are present from an early age (Johnstone et al. 1998). Given the difficulty in differentiating juvenile males from females based on physical characteristics, sex differences were not analysed for red-tailed black cockatoos. A single blood sample was taken for each bird on each day of sampling, however some birds (n = 3) were resampled at a later date. For these birds, the two sampling values were averaged to produce a single value for each parameter. Most samples (n = 61) were taken via jugular venipuncture with the bird under general anaesthesia using 3%-5% isoflurane (Veterinary Companies of Australia, Kings Park, New South Wales, Australia) with oxygen delivered at a flow rate of 1.5L/min. Induction for anaesthesia was achieved using a face mask, followed by endotracheal intubation for maintenance. Some Carnaby s cockatoos (n = 14) and forest red-tailed black cockatoos (n = 4) were manually restrained for blood collection from the medial metatarsal vein. A 23G needle and 3ml plastic syringe 160

187 were used to collect the blood sample which was used to make a smear, and then placed into a 1ml lithium heparin MiniCollect tube (Greiner Bio-One, Frickenhausen, Germany). Samples were kept cool at approximately 4 C and transported within 12 hours to Vetpath Laboratory Services, Ascot, Western Australia for processing. Haematologic analysis was performed on a Cell-Dyn 3500 automated hematology analyzer (Abbott Diagnostics, Abbott Park, Illinois, USA), run in resistant red cell mode (Walberg 2001). Red blood cell count (RBC), white blood cell count (WBC), haemoglobin (Hb), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH) and mean corpuscular hemoglobin concentration (MCHC) were measured. Packed cell volume (PCV) was measured using standard hematocrit methods after centrifugation in a Hettich Zentrifuge Haematocrit 20 (Hettich, Tuttlingen, Germany) for 3 minutes at 7500 rpm. MCV, MCHC and MCH were calculated using standard formulas from RBC, PCV and Hb concentrations (Campbell and Ellis 2007). When insufficient blood volume was available for the automated analyzer, a count was performed manually on a blood smear. Blood smears are used commonly to estimate avian white blood cell counts (Walberg 2001), due to the difficulty in drawing large volumes of blood from birds, and should always be submitted when sending blood for analysis. Smears were stained with Leishman s stain for a manual differential leukocyte count. Two hundred leukocytes per smear were counted and characterized as heterophils, monocytes, lymphocytes, eosinophils or basophils based on morphological characteristics. Avian thrombocytes are rarely counted in terms of absolute numbers, as they tend to clump and are easily lysed (Weiss and Wardrop 2010); therefore a qualitative evaluation of the number of thrombocytes in relation 161

188 to erythrocytes was performed. Biochemical analysis was performed on plasma using an Olympus AU 400 Automated Chemistry Analyzer (Olympus Optical Company, Tokyo, Japan). Parameters measured included creatine phosphokinase (CK), aspartate aminotransferase (AST), uric acid, amylase, glucose, betahydroxybutyrate (BOHB), sodium (Na), potassium (K), chloride (Cl), total protein, albumin, globulin, calcium (Ca) and phosphorus (P). All statistical analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, Illinois, USA). Outliers were detected using the SPSS program and eliminated on the basis of the Dixon s or range test. Some individual values (n = 53) were removed from the data set if the difference between the two highest (or lowest) was greater than one third of the range of all values (Solberg 1994; Thrall et al. 2004). The normality of the data for each species was tested using a Shapiro-Wilk test. For those data with a normal distribution, the mean and standard deviation of analytes were computed. For data with a non-parametric distribution, the median and percentiles were computed. The sample sizes for each species were fewer than 40 animals, so a range of values (minimum to maximum) was given as reference intervals, rather than two standard deviations (Thrall et al. 2004). Potential differences were evaluated between male and female Carnaby s cockatoos, using t-tests for normally distributed data, or Mann- Whitney tests for non-parametric data (Conover 1980; Motulsky 1995). Sexrelated differences were not assessed for the other species due to small sample sizes or lack of available data. For Carnaby s cockatoos, where the sample sizes were large enough, differences were also evaluated between birds from which blood was collected using different methods (from the jugular vein under anaesthesia or the metatarsal vein under manual restraint). Species differences 162

189 were explored using a one-way ANOVA for normally distributed data, or a Kruskal-Wallis test for non-parametric data. Statistical significance was set at a level of P <.05. When significant differences were detected, a Bonferroni correction was made to reduce the risk of Type I errors (Motulsky 1995). 4.3 RESULTS The reference intervals determined for Carnaby s cockatoos, Baudin s cockatoos and forest red-tailed black cockatoos are listed in Tables 4.2, 4.3 and 4.4 respectively. All the cockatoos were found to have adequate numbers of thrombocytes. Significant differences were detected between male (n = 13) and female (n = 21) Carnaby s cockatoos for amylase (P < 0.01) and Ca (P < 0.01). For Carnaby s cockatoos, sampling method was investigated as a factor affecting blood parameters. Several analytes (PCV, lymphocyte relative count, monocyte relative and absolute count, AST, uric acid, cholesterol, Ca and P) differed statistically between the group of birds which were anaesthetised for jugular venipuncture and those that were restrained manually for blood collection from the medial metatarsal vein (Table 4.5). There were significant interspecies differences in blood values including Hb (P = 0.04), PCV (P < 0.01), MCHC (P = 0.01), MCH (P < 0.01), MCV (P < 0.01), WBC (P = 0.01), heterophil relative count (P = 0.01), lymphocyte relative count (P = 0.01), eosinophil relative count (P < 0.01), eosinophil absolute count (P < 0.01), AST (P = 0.01), amylase (P < 0.01), glucose (P < 0.01), cholesterol (P < 0.01), albumin (P < 0.01), albumin:globulin ratio (P < 0.01) and P (P < 0.01). 163

190 Table 4.2 Values for haematologic and biochemical tests for Carnaby s cockatoos. Analyte Sex n Mean SD Median 10-90% Minimum Maximum Hb (g/l) All PCV (L/L) All RBC (x10 12 /L) All MCHC (g/l) All MCH (pg) All MCV (fl) All WBC (x10 9 /L) All Heterophils (%) All (x10 9 /L) All Lymphocytes (%) All (x10 9 /L) All Monocytes (%) All (x10 9 /L) All Eosinophils (%) All (x10 9 /L) All Basophils (%) All (x10 9 /L) All CK (U/L) All AST (U/L) All Uric acid (mmol/l) All Amylase (U/L) Males Females Glucose (mmol/l) All BOHB (mmol/l) All Na (mmol/l) All K (mmol/l) All Na/K ratio All Cholesterol (mmol/l) All Total protein (g/l) All Albumin (g/l) All Globulin (g/l) All A/G Ratio All Ca (mmol/l) Males Females P (mmol/l) All

191 Table 4.3 Values for haematologic and biochemical tests for Baudin s cockatoos. Analyte n Mean SD Median 10-90% Minimum Maximum Hb (g/l) PCV (L/L) RBC (x10 12 /L) MCHC (g/l) MCH (pg) MCV (fl) WBC (x10 9 /L) Heterophils (%) (x10 9 /L) Lymphocytes (%) (x10 9 /L) Monocytes (%) (x10 9 /L) Eosinophils (%) (x10 9 /L) Basophils (%) (x10 9 /L) CK (U/L) AST (U/L) Uric acid (mmol/l) Amylase (U/L) Glucose (mmol/l) BOHB (mmol/l) Na (mmol/l) K (mmol/l) Na/K ratio Cholesterol (mmol/l) Total protein (g/l) Albumin (g/l) Globulin (g/l) A/G Ratio Ca (mmol/l) P (mmol/l)

192 Table 4.4 Values for haematologic and biochemical tests for forest red-tailed black cockatoos. Analyte n Mean SD Median 10-90% Minimum Maximum Hb (g/l) PCV (L/L) RBC (x10 12 /L) MCHC (g/l) MCH (pg) MCV (fl) WBC (x10 9 /L) Heterophils (%) (x10 9 /L) Lymphocytes (%) (x10 9 /L) Monocytes (%) (x10 9 /L) Eosinophils (%) (x10 9 /L) Basophils (%) (x10 9 /L) CK (U/L) AST (U/L) Uric acid (mmol/l) Amylase (U/L) Glucose (mmol/l) BOHB (mmol/l) Na (mmol/l) K (mmol/l) Na/K ratio Cholesterol (mmol/l) Total protein (g/l) Albumin (g/l) Globulin (g/l) A/G Ratio Ca (mmol/l) P (mmol/l)

193 Table 4.5 Values for analytes that differed significantly between sampling method (jugular vein under anaesthesia vs. metatarsal vein under manual restraint) for Carnaby s cockatoos. Analyte t Mann- Whitney U P Anaesthesia (n) Mean ± SD (range) PCV (L/L) ± 0.03 ( ) Lymphocytes (%) ± 5.18 (5 23) Monocytes (%) ± 1.81 (1 7) Monocytes (x10 9 /L) 3.47 < ± 0.23 ( ) AST (U/L) ± ( ) UA (mmol/l) 15 < ± ( ) Cholesterol (mmol/l) ± 2.7 ( ) Ca (mmol/l) ± 0.11 ( ) P (mmol/l) ± 0.43 ( ) Manual restraint (n) Mean ± SD (range) ± 0.03 ( ) ± 8.92 (11 35) ± 1.67 (0 5) ± 0.2 (0 0.6) ± ( ) ± ( ) ± 4.30 ( ) ± 0.1 ( ) ± 0.17 ( ) 167

194 4.4 DISCUSSION This study presents the first published reference values for three species of endangered Australian black cockatoos. Previously there were only limited data available for yellow-tailed black cockatoos (C. funereus) (n = 7; n = 6) (Polo et al. 1998; ISIS 2002a) and red-tailed black cockatoos (n = 10) (ISIS 2002b). This study found statistically significant sex-based differences in amylase and calcium in the Carnaby s cockatoos. Calcium levels were significantly lower in female Carnaby s cockatoos than in males. In other species, seasonal changes in plasma calcium in females are caused by the transport of calcium-bound yolk proteins to the ovary (Harr 2002) or eggshell development (Lierz and Hafez 2005). Significantly higher levels of calcium have been found in female budgerigars than in males (Hochleithner 1989), although seasonal differences were not explored. Investigation of calcium levels across seasons would be required to determine if such seasonal changes occurred in Carnaby s cockatoos. Age-specific values are also helpful as there are often lower red cell counts, haemoglobin concentration and packed cell volume in young birds compared with adults (Jain 1993), however age is difficult to assess in birds past sexual maturity (Winker 2000) and life histories of wild birds are often unknown. A previous publication on African grey parrots (Psittacus erithacus) also found significantly higher amylase levels in males than in females (Hochleithner 1989), however, as in this study, this is not easily explained and interpretation of amylase levels in birds can be difficult (Hochleithner 1994; Ferrer and Dobado-Berrios 1998; Zaias et al. 2000; Lierz and Hafez 2005; Gelis 2006). For individual 168

195 cockatoos there may not be a significant difference in values for males and females (as for calcium), however further investigation is required to explain these results, which may be due to factors that were not included in this study, or may be rendered less significant with a larger sample size. Differences between the two groups of Carnaby s cockatoos from which blood was collected using different methods should also be evaluated with caution given the small sample sizes. The differences could be due to the different venipuncture sites, but if representative of the population, they are more likely to be influenced by the presence or absence of gaseous anaesthesia. The effect of venipuncture site has been demonstrated in reptiles (Gottdenker and Jacobson 1995; Lopez-Olvera et al. 2003), but these are likely to be due to lymph contamination, which does not occur with avian venipuncture. However, while changes in analytes with anaesthesia in birds have not been closely examined, isoflurane has been shown to have some effect on uric acid and potassium levels in American kestrels (Dressen et al. 1999). This may be due to reduced hepatic synthesis or renal excretion of uric acid under anaesthesia. Other factors that may be important when considering differences in restraint technique include fasting periods prior to anaesthesia and increased stress due to prolonged handling associated with conscious restraint (Scope et al. 2002). In this study, the Carnaby s cockatoos that were manually restrained had a higher mean PCV than those that were anaesthetised. This may have been associated with the stress from handling, as slight increases in the red blood cell mass in birds have been associated with excitement or stress (Fudge 1984). The other differences among parameters in the two groups of Carnaby s cockatoos are not easily explained. 169

196 One of the most significant interspecies differences was in the eosinophil count. Forest red-tailed black cockatoos had a median eosinophil relative count of 4% of the total leukocyte count (with a range of 0% 16%) which was statistically higher (P <0.01) than Carnaby s cockatoos and Baudin s cockatoos (which both had median values of 0% and ranges of 0% 1%). These differences have been noticed by clinicians but have not previously been statistically confirmed. Eosinophils are irregularly round granulocytes, with a lobed nucleus (see Figure 4.1), typically less common in avian blood than heterophils (Clark et al. 2009). In some species such as parrots, eosinophils may have round, light blue granules, whereas in other species the granules are red-orange (Latimer and Bienzle 2010). Due to the potential for eosinophils and heterophils to be misidentified by automated analysers, blood smears should always be examined in every bird, to ensure good correlation between the automated and manual count. Eosinophils are reportedly uncommon in psittacine blood (Tell and Citino 1992; Polo et al. 1998) unlike raptors, waterfowl and herons (Feldman et al. 2000; Fudge and Joseph 2000). The function of the avian eosinophil remains unclear, but elevated counts have been associated with delayed (type IV) hypersensitivity (Thrall et al. 2004) and gastrointestinal parasitism (Samour and Howlett 2008). The cause of the higher eosinophil count in forest red-tailed black cockatoos is unknown and may just be a species-related difference. However, another possible explanation could be differences in parasite infestation which has not been investigated closely in black cockatoos. Variations in circulating leukocyte counts have been found in birds with higher blood parasite infestation compared with those with lower infestations (Fokidis et al. 2008) and in tropical species as compared with closely related non-tropical species (Moller 1998). 170

197 Figure 4.1 A normal eosinophil and erythrocytes in the peripheral blood film from a parrot (Wright s stain, 1000X) (Campbell and Ellis 2007). There is likely to be reasonable genetic diversity among the cockatoos used for this study, as those birds from the rehabilitation centre originated from different wild populations. However it is important to note that the birds used for this study had been captive for at least 12 months and there may be disparity in blood values in wild conspecifics. Variations in blood parameters have been demonstrated in wild and captive birds (Wolf et al. 1985; Lewandowski et al. 1986; Dobado- Berrios et al. 1998; Lashev et al. 2009), which may be attributable to differences in diet and physical activity. However, there are difficulties inherent in the compilation of reference values from free-living birds, as there are often only small sample sizes available and capturing free-flying birds can cause significant amounts of stress and pose risks to the safety of the animals. 171

198 CHAPTER FIVE: RETENTION OF TRANSMITTER ATTACHMENTS FOR BLACK COCKATOOS (CALYPTORHYNCHUS SPP.) This chapter is modified from the following paper: Le Souëf A, Stojanovic D, Burbidge A, Dawson R, Heinsohn R, Vitali S, Warren K Retention of transmitter attachments for black cockatoos (Calyptorhynchus spp.). Pacific Conservation Biology 19(1):

199 5.1 INTRODUCTION Of the 355 extant psittacine species recognised by BirdLife International, 39% are listed as being near threatened to critically endangered, making parrots one of the most threatened groups of birds (BirdLife International 2011). Habitat alteration is the major threat, impacting a broad range of species, including several species of Australian black cockatoos (Calyptorhynchus spp.) (Forshaw and Knight 2010; Garnett et al. 2011). Despite extensive research into some aspects of their life histories, a lack of knowledge about habitat use and landscape-scale movements of black cockatoos impedes effective management strategies for most species (Davies 1966; Saunders 1980; Cameron 2006; Chapman 2007). There is also a paucity of knowledge on the post-release survival of rehabilitated black cockatoos, which could be improved upon with the use of telemetry to track released birds. Telemetry units have been used to provide detailed information on the movements, habitat use and life history of many bird species, including some parrots (Elliott et al. 1996; Meyers 1996; Myers and Vaughan 2004; White et al. 2005), however the powerful bills and allopreening behaviour of larger species like cockatoos may be a major impediment to retention of external devices (Saunders 1979a, 1980; Nelson and Morris 1994; Higgins 1999; Cooper et al. 2002; Cameron 2006). Further, externally attached transmitting devices can impose physiological or behavioural costs on the study subject (Hines and Zwickel 1985; Gessaman and Nagy 1988; Massey et al. 1988; Obrecht et al. 1988; Longcore et al. 2000; Iverson et al. 2006; Vukovich and Kilgo 2009), so evaluating transmitter effects is important in advance of field deployment. 173

200 Here, the results of a trial of six transmitter types and three attachment methods on captive black cockatoos to investigate the feasibility of deploying transmitters on wild black cockatoos are reported. 5.2 METHODS Study area and species Seventeen Carnaby s cockatoos (Calyptorhynchus latirostris), six Baudin s cockatoos (C. baudinii) and six forest red-tailed black cockatoos (C. banksii naso) (Table 5.1) at the Kaarakin Black Cockatoo Rehabilitation Centre in Martin, Western Australia were used for the trial. Most birds used (n = 22) were capable of normal flight. All birds were housed in three aviaries (two were 12m x 5m x 2.5m high and one was 6m x 3.5m x 2.5m high) fitted with hollow logs and artificial nest boxes. Seed, water and native browse were available ad libitum. Table 5.1 Number of black cockatoos of each species assigned to different transmitter groups. Transmitter type Carnaby s cockatoo Baudin s cockatoo Forest red-tailed black cockatoo Collar Tail-mount (6g) Tail-mount (12g) Harness (12g) Harness (16g) Harness (20g)

201 5.2.2 Transmitter attachment Two tail-mounted radio Platform Transmitter Terminals (PTTs) weighing 6g (n = 6) and 12g (n = 3) (Sirtrack Wildlife Tracking Solutions, Havelock, New Zealand), three harness-mounted solar PTTs weighing 12g (n = 4), 16g (n = 4) and 20g (n = 5) (North Star Science and Technology, Virginia, USA); and a radio collar weighing 24g (n = 7) (Holohil Systems Ltd., Ontario, Canada) were tested. Transmitters were fitted to the birds under isoflurane gaseous anaesthesia (Figure 5.1), to reduce stress from handling. The maximum weight of the transmitters was 4% of body mass, which is below the maximum weight typically used in birds (Gessaman and Nagy 1988; Kenward 2001; Godfrey et al. 2003). 175

202 Figure 5.1 Field anaesthesia set up for transmitter attachment. An X-attachment method (Buehler et al. 1995) was modified for the harness design, incorporating a single weak link (Figure 5.2). Neoprene pads were glued to the base of the PTTs to reduce abrasion and to raise the solar panels above the feathers. Teflon ribbon (Telonics Inc., Mesa, Arizona, USA) was used to form the harness and position the unit between the scapulae so that cranial and caudal movement was limited to 0.5 cm (Buehler et al. 1995). Harnesses were secured with absorbable suture material where the four ribbons met at the sternum (the weak link ) and sealed with glue to prevent fraying. 176

203 Figure 5.2 Harness design for black cockatoos showing modified Xattachment method (the central cross rests on the sternum of the bird). The 12g tail-mounted PTT was attached dorsally to the two central feathers using nylon fishing line and cyanoacrylate glue and the antenna was tied and glued along its length to one tail feather. The 6g tail-mounted PTT was attached in the same way but to the ventral surface of the tail feathers (Figure 5.3). 177

204 Figure 5.3 6g tail-mounted PTT attached to ventral tail feather. Collars were attached with the barrel resting above the sternum and bolted in the middle tightness setting, with approximately 1cm of space between the collar and the neck (Figure 5.4). This fit allowed for filling of the crop without interference and avoided restriction of the airways, but was not too loose as to present a significant snagging risk with the lower mandible (Bjork 2004). The collars used are designed to detach from the bird after the bolt oxidises and falls away. 178

205 Figure 5.4 Collar attached to cockatoo with barrel resting above the sternum. Retention time was defined as the point where the transmitter package was either damaged enough to impair transmission or was no longer attached to the bird. At the end of the study (554 days) any transmitters were removed that had not already failed and differences in mean retention time were compared between transmitter packages. Physical examinations of feathering, skin and general condition were carried out one week following attachment, then at monthly intervals for 12 months, to check for any physical effects on the birds from the transmitters. 179

206 5.2.3 Statistical analyses Analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, Illinois). Data were tested for normality using a Shapiro-Wilk test. A Kruskal-Wallis test was used to compare retention times between groups, using transmitter attachment method (collar, tail-mount or harness) as the grouping variable and retention time (days) as the dependent variable. Transmitter package (collar, 6g tail-mount, 12g tail-mount, 12g harness, 16g harness or 20g harness) was also used as the fixed factor with retention time as the dependent variable. When a significant difference (P < 0.05) was detected a Mann-Whitney test was used to make pairwise comparisons. 5.3 RESULTS In this study, retention times varied significantly among attachment method (ie collar, harness, tail-mount) (χ 2 (2) = 7.14, P = 0.026). Collars were retained significantly longer than tail-mounts (P = 0.027) and harnesses (P = 0.046), with no statistical difference between the mean retention times of tail-mounts and harnesses (P = 0.126). There were no significant differences in retention times among the six different transmitter packages (χ 2 (5) = 7.71, P = 0.17) (Table 5.2). 180

207 Table 5.2 Transmitter retention times (days) for collar, 12g tail-mount, 6g tail-mount, 12g harness, 16g harness and 20g harness transmitter packages on black cockatoos. Retention time (days) Transmitter type n Range x S.E. Collar Tail-mount (12g) Tail-mount (6g) Harness (12g) Harness (16g) Harness (20 g) There was no evidence of skin irritation or feather wear underneath the transmitter attachments at the monthly physical checks. Causes of transmitter failure varied with transmitter type. Harnesses failed due to damage inflicted by cockatoo (n = 7), spontaneous detachment from the cockatoo (n = 4) and manual removal due to a loose thread (n = 1). All collars remained attached for the study, except for one that was removed immediately by the authors after a Baudin s cockatoo entangled its lower bill in the collar. Tailmounts failed due to chewing by the cockatoo (n = 2) or spontaneous detachment (n = 7). 181

208 5.4 DISCUSSION These data suggest that some telemetry attachments may be suitable for large psittacine birds with strong bills. In particular, tail-mounted attachments are suggested for telemetry studies, as they are likely to present the lowest snagging risk to the bird, while still having retention periods in our study of up to 287 days. Snagging of harnesses was not observed, but this attachment method has been associated with higher risks than tail-mounted packages (Kenward 2001; Peniche et al. 2011). The use of gaseous anesthesia (such as isoflurane) is also recommended during transmitter attachment, as during this study it enabled the correct fitting of the transmitter packages and minimised handling stress. Retention times for transmitters were highly variable and in part, reflect individual reactions of birds to the attachments. Cockatoos damaged nine of 29 transmitter units and such attrition rates should be anticipated when planning tracking studies on these birds. Some birds seemingly ignored their attachments, with one redtailed black cockatoo laying and incubating an egg in a hollow log while wearing a harness-mounted PTT. Species differences may have had an effect on the results of the study, however numbers of each species were limited to the availability of non-release birds at the time. Although the sample size precluded statistical comparison between species, red-tailed black cockatoos (which have considerably larger bills than the other species) were noted to inflict the greatest bite damage to both harness and tailmounted packages. Also, the longer, slender bill shape of the Baudin s cockatoo may predispose this species to entanglement with collars, which occurred with one bird in this study. 182

209 At the end of the study, the detachment mechanism of the collars was simulated, which is designed to occur when the bolt oxidises, by removing the bolts securing the collars. Following this, one collar detached from a bird within 24 hours; the remaining collars also detached spontaneously, up to six months later. However, in the field, the detachment mechanism depends upon the disintegration of the main bolt attachments for detachment. Design modification to collars may be necessary to ensure cockatoos are eventually freed from their transmitters, especially in dry climates such as in Australia where the bolt is unlikely to rust quickly. Testing transmitter attachments on captive birds has the advantage of allowing detailed observation of transmitter retention. This study is a first step towards further research into the habitat preferences and movements of black cockatoos using telemetry. It is also important due to the need for post-release studies to appraise the long-term success of black cockatoo rehabilitation. 183

210 CHAPTER SIX: THE DEVELOPMENT OF AN AGING CURVE FOR BLACK COCKATOOS USING PENTOSIDINE ANALYSIS 184

211 6.1 INTRODUCTION Aging in birds The process of aging is progressive, universal and deleterious (Strehler 1962). It involves an assortment of processes, including free radical production and resulting oxidative damage, progressive erosion of telomeres and cellular senescence, and trade-offs in hormone signalling pathways and immunosenescence, leading to an increased risk of infection and autoimmune disease (Vleck et al. 2007). Many bird species live for paradoxically longer time periods compared to other animals, despite their relatively high metabolic rate, body temperature and blood glucose level (Holmes et al. 2001). Theoretically, these attributes should contribute to increased rates of aging due to accelerated oxidative changes and the Maillard reaction (Harman 1956; Monnier 1990). As in mammals, many species of birds show an inverse relationship between longevity and body mass, however birds can live up to three times longer than a mammal of the equivalent body mass (Lindstedt and Calder 1976). Birds also show slow rates of reproductive aging compared to other animals, with some avian species actually showing an increase in average reproductive success in older breeding pairs (Pugesek 1981; Holmes and Ottinger 2006). Explanations for how birds achieve their unusually long life spans in the face of high metabolic rates and other contradictory factors have proved intangible (Ricklefs and Cadena 2008). Recent discoveries have revealed that birds have adaptations to resist senescent cellular changes, which has led to great interest in 185

212 the use of birds as models for human aging studies (Holmes and Ottinger 2003; Holmes and Ottinger 2006). However, avian aging defence mechanisms are complex and not completely understood. They include neural regeneration and superior defences against oxidative damage at a cellular level, particularly against reactions that involve harmful byproducts of glycoxidation reactions (Ogburn et al. 1998; Scharff et al. 2000; Ogburn et al. 2001). Avian antioxidant defences may include uric acid, which is found in high levels in birds (Klandorf et al. 1999). Birds have also been shown to accumulate advanced glycosylation end products (AGEs), such as pentosidine, at a slower rate than mammals (Iqbal et al. 1999). Further, birds have high levels of saturated membrane fatty acids, which have been implicated as being protective against oxidative damage (Hulbert 2008). Telomere length has also been a focus of research on avian aging, as the rate of change in telomere length has been correlated with life span in both birds and mammals (Vleck et al. 2003). Individuals and species with telomeres that shorten at a slower rate may exhibit longer life spans because of the protective actions of longer telomeres against chromosomal fusion and cell replicative senescence, suggesting that regulation of telomere length and life span have co-evolved (Blackburn 2000; Pauliny et al. 2006). Three species with exceptionally long life spans, the Leach s storm petrel (Oceanodroma leucorhoa), the wandering albatross (Diomedia exulans) and the European shag (Phalacrocorax aristotelis) actually exhibit a lengthening of telomeres over time, possibly by upregulation of telomerases (Haussmann et al. 2003; Hall et al. 2004). However, other long-lived species such as the great frigatebird (Fregata minor) have shown a poor relationship between telomere length and age (Juola et al. 2006). More research is 186

213 needed to further characterise a possible relationship between telomere length and longevity in birds. Other explanations for exceptional longevity in avian species link to evolutionary theories of longevity (Munshi-South and Wilkinson 2010); for example, the ability to fly reduces the vulnerability of birds to environmental factors and may contribute to a longer life span (Holmes and Austad 1994). Lower extrinsic mortality rates are thought to expose late-acting mutations to purifying selection (Holmes and Austad 1995). Another theory associates longevity with migration or tropical residence, with benign environments resulting in a reduction of extrinsic mortality and resting metabolic rates, and therefore long life span (Wikelski et al. 2003; Moller 2007). There are a wide range of life spans and aging patterns amongst avian species. Parrots (Psittaciformes), seabirds (Charadriiformes), songbirds (Passeriformes), hummingbirds (Apodiformes) and raptors (Falconiformes) all include within their order species with long life spans and slow rates of aging in relation to their body size, while poultry (Galliformes) are the shortest lived and most quickly aging birds (Holmes and Austad 1995). Long life spans may have been critical in the evolution of family living in birds, because longevity favours delayed reproduction and gives parents the opportunity to make prolonged investments in their offspring, an option not available for short-lived species (Covas and Griesser 2007). Parrots show a positive relationship between communal roosting and longevity, due to improved detection of scarce resources and reduction of predation through increased vigilance (Westcott and Cockburn 1988; Jullian and Clobert 2000; South and Pruett-Jones 2000; Munshi-South and Wilkinson 2006). 187

214 There is also an association between longevity and diet type, where granivorous parrots were found to live longer than frugivorous, nectarivorous or omnivorous parrots (Munshi-South and Wilkinson 2006). However, this may also be related to the comparative ease with which natural diets can be replicated for captive granivorous birds, compared with diets based on fruit and nectar. The provision of an unbalanced diet to frugivorous or nectarivorous birds in captivity may lead to chronic stress and therefore a reduction in reported life spans (Munshi-South and Wilkinson 2006) Longevity in black cockatoos In captivity, cockatoos have the greatest longevities and reproductive life spans of all psittacine birds, with larger cockatoos generally living longer than smaller species (Brouwer et al. 2000). There are literature reports of a Baudin s cockatoo having reached at least 47 years of age, and a red-tailed black cockatoo at London Zoo that lived for 45 years (Jupp 1996; Brouwer et al. 2000). Captive cockatoos have remained reproductively active at more than 30 years of age in a number of institutions (Dolan and Moran 1970; Muller 1975). The average life spans of wild birds are usually thought to be shorter compared with their captive conspecifics, presumably due to increased energy expenditure required by wild birds for the maintenance of life and natural causes of mortality such as disease, predation and accident (Müller et al. 2010). However, life span data of wild animals are often not available, particularly for free-flying birds that are generally difficult to recapture or resight. Prior to this study, data on maximum and reproductive ages of wild black cockatoos have resulted from 188

215 resightings of stainless steel patagial tags or leg bands, attached during extensive ecological studies on the Carnaby s cockatoo in the 1970s to mark individuals for ongoing observation (Saunders 1982b). Monitoring of a population at Coomallo Creek over 28 years confirmed that one female reached 19 years of age (Saunders and Ingram 1998). In 2008, a stainless steel patagial tag was found on the ground, having come from a Carnaby s cockatoo that was likely to have been recently predated (Saunders and Dawson 2009). Records show that this cockatoo was at least 34 years old at the time of death, which remains a record for longevity for this species in the wild Age estimation and cockatoo conservation Age data are important to conservation biologists, wildlife managers and veterinarians, because age plays a role in survival rates, reproductive success, and disease susceptibility among individuals and within wild animal populations (László et al. 2005; Fallon and Klandorf 2009). The ability to reliably determine age in birds could fill an important role in species conservation efforts and the pairing of endangered species in captive breeding programs (Chaney et al. 2003). The age demographics of wild populations of black cockatoos, including the percentage that remain within reproductive age, are largely unknown. Due to the loss of large areas of breeding habitat following clearing of inland regions of south-west Australia last century, there may be suboptimal rates of juvenile recruitment. Although reasonable numbers of cockatoos remain throughout wide areas of their habitat, many of these remaining birds may be of post-reproductive age, which could result in a rapid reduction in numbers after these older birds die. If the age demographic of wild populations can be determined, it would inform 189

216 conservation management decisions such as targeted captive breeding for the release of birds of a certain age. The capacity to estimate the age of sick and injured black cockatoos presented to the PZVD would provide valuable information on the life histories of cockatoos in the rehabilitation program. Age estimation of these birds would also be helpful when considering them as suitable candidates for captive breeding Methods of age determination in birds Anatomical features that indicate advanced age in mammals, such as increased molar wear and wrinkled skin, do not exist in birds. The anatomy of birds, specifically the development of a beak in place of teeth, feathers covering the skin, and pneumatic bones (which do not show obvious signs of aging, such as reduction in density, as do non-pneumatic bones) complicate the ability to determine the age of birds from external characteristics past sexual maturity (Winker 2000; Chapman et al. 2005). Historically, estimations of age in wild birds have been based either on rare longterm studies on marked individuals, or in-hand measurement of birds as part of banding studies, such as body morphometrics, and moult and plumage criteria (Prince and Rodwell 1994; Mather and Esler 1999). However, these techniques are limited in their ability to provide reliable data. Other methods include measurement of telomere rate of change (see Chapter 6.1.1), however this technique requires further work to determine its suitability for reliably aging birds. 190

217 Pentosidine measurement The limitations of banding data and other methods of age estimation in birds and the value of avian age information, has led to the recent development of a technique for estimating the age of adult birds using pentosidine analysis (Fallon and Klandorf 2009). Pentosidine is a fluorescent, irreversibly synthesised compound comprising a lysine and an arginine residue, cross-linked by a pentose (Sell and Monnier 1989). It is one of the AGEs produced during a sugar-protein interaction called non-enzymatic glycation (Suji and Sivakami 2004). The accumulation of pentosidine in collagen has been found to correlate with chronological age in mammals and birds (Sell et al. 1996; Iqbal et al. 1999). It accumulates at a predictable rate, which combined with its fluorescent properties makes it a reliable and measurable biomarker for aging (Chaney et al. 2003; Fallon et al. 2006b). By taking skin samples from a range of birds of known ages and measuring concentrations of pentosidine within the collagen, age predictions can be made for conspecifics for which life histories are not known. Sampling methods for this technique are also attractively straightforward; a safe and effective protocol for taking skin biopsies from conscious wild birds has been developed (Cooey et al. 2010). The technique has been proven for use in wild birds, including psittacines (Klandorf et al. 2000), however a comprehensive aging curve for a species similar to black cockatoos did not previously exist. Repeatability of the test has also been proven by testing pentosidine in duplicate samples from 224 birds, which produced no significant difference between samples (P = 0.147) (C. Cooey, unpublished data). 191

218 The concentration of pentosidine differs according to the location from which it is taken, therefore care should be taken to consistently sample the same site in order to avoid conflicting measurements (Chaney et al. 2003). Also, there may be slight variability in pentosidine concentrations due to differences in diet, metabolic status, habitat or sex. However, a correlation still exists between pentosidine and age when these factors are considered in the model. Further research is required to determine the effect of physiology and environmental conditions on pentosidine accumulation. 6.2 AIM The aim of this chapter was to create a tool for the age estimation of black cockatoos (Calyptorhynchus spp.) via measurement of pentosidine levels in skin samples of captive, rehabilitated and nestling cockatoos of known age. 6.3 MATERIALS AND METHODS Sample procurement Skin samples were collected from 53 black cockatoos from the genus Calyptorhynchus: Carnaby s cockatoos (C. latirostris) (n = 28), Baudin s cockatoos (C. baudinii) (n = 6), forest red-tailed black cockatoos (C. banksii naso) (n = 9), south-eastern red-tailed black cockatoos (C. banksii graptogyne) (n = 4), inland red-tailed black cockatoos (C. banksii samueli) (n = 4) and yellow-tailed black cockatoos (C. funereus) (n = 2). Seven of these samples were taken from wild black cockatoo nestlings (under a DEC Regulation 17 Licence), while the 192

219 rest were taken from cockatoos held in Australian zoos, private collections and rehabilitation centres (Table 6.1). Table 6.1 Origins of the black cockatoos from which skin samples were taken for pentosidine analysis. Species Wild (n) Rehabilitation Centre (n) Zoo (n) Private collection (n) C. latirostris C. baudinii C. banksii naso C. banksii graptogyne C. banksii samueli C. funereus Total Actual age was determined for nestlings using a folded left wing measurement (Saunders 1982b) and was determined for adult birds by reviewing records of hatch dates. Cockatoos were only included if their hatch date was known or if they had entered captivity as a juvenile, in which case their age was known within one year. The age range of the cockatoos was 6 weeks to 33 years (Table 6.2). 193

220 Table 6.2 Ages of black cockatoos from which skin samples were taken for pentosidine analysis. Age group Number of black cockatoos sampled (n) 0-5 months months years years years years years years years Sample collection Samples were collected from cockatoos under gaseous isoflurane anaesthesia, or, in the field, when cockatoos could be handled with minimal stress, samples were collected under manual restraint. A skin biopsy was taken from the inner patagium, where there was minimal vasculature (Figure 6.1). Any feathers obscuring the site were removed and the skin was prepared with a mixture of chlorhexidine and 70% ethanol. A subcutaneous injection of 0.05ml lignocaine (diluted 1:1 with 0.9% sodium chloride solution) using a 25G needle and 1ml syringe was used to provide local anaesthesia to the biopsy site, and to raise the thin epidermal layer from the underlying tissue. The skin was stretched so that it was taut and a disposable 6mm diameter biopsy punch (Paramount Surgimed Ltd, 194

221 New Delhi, India) was used to penetrate the epidermal layer. Atraumatic forceps and scissors were used to remove the skin sample and place it into an Eppendorf tube with distilled water. The tube was placed into the freezer, or if in the field, was placed on ice until it could be frozen later in the day. After haemostasis was achieved, Vetbond Tissue Adhesive (3M, St Paul, Minneapolis, USA) was used to seal the defect. The sampling process was then repeated on the other patagium. Duplicate samples were taken (i.e. one sample from each wing), in case samples were lost or damaged during transportation to the United States. a b c d Figure 6.1 Collection of the 6mm skin biopsy from the inner patagium of a black cockatoo (a: injecting lignocaine; b: making the biopsy incision; c: sealing the incision with tissue adhesive; d: the final appearance). 195