The true incidence of traumatic spinal cord injuries See paper by Sabre et al. on page 768. doi:10.1111/ene.12540 The incidence of traumatic spinal cord injuries (SCIs) is usually based on the number of patients arriving at hospital alive. The true incidence, however, should include patients who died at the scene. Very few stud- ies have managed to include all patients [1–5]. Sabre et al. have collected data of every traumatic SCI in Estonia during 2005–2007 [1]. By ascertaining the numbers though medical records and autopsy reports from every hospital in the country, they calculated the true incidence of traumatic SCI for the period. The results are astonishing. Sabre et al. found an annual incidence rate of 93.9 and 43.8 per million pop- ulation respectively when including and excluding pre- hospital deaths [1]. Less prominent differences have been found in Mississippi [2] and Alberta [3]. The inci- dence rate in Mississippi was 77 and 59 SCI per million population per year respectively when including and excluding pre-hospital deaths [2]. In Alberta, Canada, the overall annual incidence rate was 52.5 and 44.3 million per population per year respectively [3]. Fifty-three per cent died prior to hospitalization in Estonia [1], compared with 37.7% in Minnesota, USA [4], 16.0% in Portugal [5] and 15.8% in Alberta, Canada [3]. There were striking differences between the victims who died pre-hospital and the patients who were alive upon arrival with regard to age, gender, level of injury and aetiology. Persons who died at the scene of the accident were 10 years younger than those who died later [1]. Sixty-five per cent of all women died before arrival at hospital, compared with 50.2% of men [1]. No other publication has previously reported this. The highest mortality was amongst the patients with multiple trauma, 86.1% [1]. These findings corre- late with a Swedish study where the patients who died before admission to hospital had a much higher injury severity score compared with those who were alive upon arrival [6]. High cervical injuries are the most serious injuries and 67.8% of the patients with these injuries died [1]. Motor vehicle accidents (MVAs) accounted for 53.2% of injuries, followed by falls 27.4% [1]. Only 25.0% of all victims of MVAs were alive upon arrival at hospital [1]. As a result the main cause of traumatic SCI in Estonia when including pre-hospital deaths is MVAs and when excluding pre-hospital deaths falls [1]. These findings are supported by the Swedish study which found a higher percentage of pre-hospital deaths due to MVAs than falls [6]. The majority of injury-related deaths occurred either at the scene or before arrival at hospital [6]. Traffic accidents were the leading cause of death amongst people aged 15–64 years in Estonia in the period 2000–2002, based on post-mortem reports, and 78.3% of the victims were men [7]. Fifty-eight per cent of the victims were car occupants and 31% were pedestrians [7]. Alcohol was a major contribut- ing factor in these accidents, and the percentage of alcohol-related deaths was highest amongst pedestri- ans [7]. In the present study 55.2% of the population and 46.8% of those who died had consumed alcohol prior to the injury [1]. Studies have shown that both male gender and age are associated with increased mortality rates amongst injured persons [8]. The high mortality among women as found by Sabra et al. [1] is interesting and calls for further studies. Collecting true epidemiological data on trauma is a difficult exercise but has important implications with regard to the development of prevention strategies, locally, regionally and nationally. The meticulous work that Sabre et al. have done proves the necessity of including pre-hospital deaths in future epidemiolog- ical studies of traumatic SCI. E. M. Hagen a,b,c a Department of Neurology, Spinal Cord Injury Centre of Western Denmark, Viborg Regional Hospital, Viborg, Denmark; b Institute of Clinical Medicine, University of Bergen, Bergen, Norway; and c Institute of Clinical Medicine, Aarhus University, Aarhus, Denmark (e-mail: ellen.merete.hagen@helse-bergen.no) References 1. Sabre L, Remmer S, Adams A, et al. Impact of fatal cases on the epidemiology of traumatic spinal cord injury in Estonia. Eur J Neurol 2015; 22: 768–772. 2. Surkin J, Gilbert BJ, Harkey HL III, Sniezek J, Currier M. Spinal cord injury in Mississippi. Findings and evalu- ation, 1992 1994. Spine 2000; 25: 716–721. © 2014 EAN 743 EDITORIAL EUROPEANJOURNALOFNEUROLOGY