INTRODUCTION Vascular injuries have been well documented during wartime but there have been few reviews of the epidemiology of vascular trauma in civilian practice. 1–3 The increase in violent crime together with the increase in high-speed motor vehicle crashes has resulted not only in an increase in vascular trauma in the USA, but also a change in the types of injury sustained. This has been most noticeable in urban settings where very high incidences of neck and trunk vascular injuries have been documented. 3 This is in marked contrast to the great predominance of extremity vascular trauma observed in military conflicts. 4 It has been noted that when compared to their urban counterparts, residents of rural areas in the USA have longer hospital stays and greater mortality rates following vascular trauma. 1 This has raised concerns relating to response times and transfer capabilities. This has particular relevance to Australia as a consequence of its size and remote rural populations. To date there has been no recent study in Australia on the epidemiology of vascular injury. The Royal Perth Hospital (RPH), although not the exclusive provider for vascular trauma, is the largest provider of rural trauma services in Western Australia, accepting 62% of all air ambulance transfers. It is also the largest provider of trauma ser- vices to the metropolitan area, accepting 34% of all land ambulance transfers. The present paper reviews the RPH experience of vas- cular trauma with respect to the epidemiology. METHODS All individuals who presented to the RPH and had a vascular injury between August 1994 and January 2000 were collected from the prospective trauma database maintained at the RPH. This database records all trauma patients admitted to the RPH for longer than 24 h and those who died in hospital. The data does not include those dying at the scene and in transit to RPH. Data are collected by research assistants who are present during trauma resuscitation, and who follow trauma patients through to their eventual dispatch from the trauma service. All collected information is entered into a custom-made database (Microsoft Access 97). Pre-hospital data were assimilated from information recorded by on-scene ambulance and paramedic personnel, as well as from history obtained from the patient on arrival in the Emergency Department. Injuries are classified using the Abbreviated Injury Score (AIS 90 code; Association for the Advancement of Automotive Medicine, Des Plaines, IL, USA). The five major vascular injury groups as defined by AIS 90 predot codes (neck (320202.3–321099.2), thorax (420202.4– 422299.2), abdomen (520202.4–521699.3), upper limb (711000.3– 713000.3; 720202.2–720806.3) and lower limb (811000.3– 813004.3; 820202.3–820899.2)) were analysed in turn. RESULTS For the 58 months between August 1994 and January 2000, 15 794 patients were entered into the RPH Trauma Registry database. ANZ J. Surg. (2001) 71, 461–466 ORIGINAL ARTICLE AN EPIDEMIOLOGICAL VIEW OF VASCULAR TRAUMA IN WESTERN AUSTRALIA: A 5-YEAR STUDY ROBIN GUPTA,* SUDHAKAR RAO* AND KISHORE SIEUNARINE † Departments of *General and † Vascular Surgery, Royal Perth Hospital, Western Australia, Australia. Background: Although Australia encounters vascular injury less frequently than countries such as the USA, the predominance of blunt trauma with associated complex injuries continues to pose problems for clinicians. The present paper reviews Royal Perth Hospital’s experience of vascular trauma epidemiology. Methods: All individuals who presented to the Royal Perth Hospital (RPH) with a vascular injury between August 1994 and January 2000 were identified from a prospective trauma database. Injuries were classified using the Abbreviated Injury Score (AIS 90 code). The five major vascular injury groups as defined by AIS 90 (neck, thorax, abdomen, upper limb and lower limb) were analysed in turn to determine which injuries were most frequent, what the common causes were, effects of alcohol and drugs, common associated injuries, investigations performed, mortality, proportion of rural vascular trauma with transfer details and lengths of hospital stay. Results: During the study period, 153 patients (1% of total trauma workload) sustained 175 vascular injuries. The commonest causes of injury for each group were stabbing (neck and lower limb), motor vehicle accident (thorax and abdomen) and work machinery (upper limb). Mortality was highest for thoracic injury (14/32; 44%) followed by abdominal injury (10/29; 34%). The ratio of blunt to penetrating vascular trauma is high and is increasing, as is the ratio of truncal to extremity vascular trauma. The most common artery–vein combination injury was to the femoral vessels. Conclusions: The injury patterns emerging from the present study will hopefully help all medical personnel to recognize the poten- tial for vascular injury in a trauma setting. Key words: epidemiology, motor vehicle accident, trauma registry, trauma team, vascular trauma. Correspondence: Mr R. Gupta, 5 Belsize Avenue, Hampstead, London NW3 4BL, UK Email: robingupta@netscapeonline.co.uk Accepted for publication 9 April 2001.