Epidemiology of paediatric trauma in Norway: a single-trauma centre observational study

Trauma is a major cause of mortality and morbidity in children globally, with road injuries having the highest incidence rates [1]. The burden of injury experiences both substantial geographical and sociodemographic differences, with the highest incidence rates in low-income countries and lowest in the industrialized countries in Western Europe [1]. Also, within countries, the burden of injury is higher among children from low-income families [2].

Though the incidence of severe trauma among the paediatric population is declining in Europe, it still represents a major public health problem in the European Union [3]. As mortality rates have declined during the last two decades, morbidity after trauma is receiving more attention [4]. Many children, who survive a major trauma, are left with disabilities that affect their development, education and social life [2]. A report by the World Health Organization (WHO) showed that nearly 50% of children under the age of 12 presenting to an emergency department with injuries were left with some form of disability [2].

In Norway, the mortality rate has dropped dramatically since the early 1970s. While 37 per 100,000 boys aged 0–17 died from accidents in 1970, the same number had decreased to 2 per 100,000 in 2012 [5]. Injury-related deaths sustained at traffic accidents were most common [5]. A study published in 2012 by Kristiansen et al. also showed substantial geographical differences in the distribution of fatal trauma with the highest figures in rural areas [6].

Even though several publications describing trauma systems and traumatic injuries have been published in the last few years, there is still a lack of detailed clinical knowledge concerning the paediatric trauma population [6‐9]. Previous Scandinavian epidemiologic studies on paediatric trauma also mainly focus on those with severe injuries, traumatic brain injuries and rates of mortality [7, 9]. No regional or national studies have been conducted on the broader segment of those potentially severely injured, which also incorporates those with potentially moderate injuries. The knowledge of paediatric trauma in Norway is mostly based on national white paper reports, which mostly contain national overall figures of patients treated and corresponding fatality rates [5, 10].

Epidemiologic baseline values are needed in order to evaluate the potential need for improved triage and preventive and rehabilitative measures. Therefore, the aim of this study is to give a detailed description of epidemiology, resource use and outcome for all potential severely injured paediatric patients being admitted to a regional trauma centre.

The study is an observational retrospective analysis of prospectively collected data from a regional trauma centre in Central Norway. The study follows the “Strengthening the Reporting of Observational Studies in Epidemiology” (STROBE) recommendations for reporting of observational cohort studies [11].

A total of 873 patients were included, of which 536 (61%) were male. Children aged 0–17 years constituted 17% (873/5013) of all patients treated by the trauma team in the study period. The median age was 13 years (IQR 7–16). Six per cent (n = 52) of the patients were transferred to the trauma centre from a local emergency hospital, with the remaining 94% (n = 821) being transported directly from the scene (Table 1). Blunt trauma was the dominating type of injury in 851 patients (98%), and traumas related to traffic (n = 532/61%) and falls (n = 233/27%) were the most common injury mechanisms. Eight patients (1%) died within 30 days of hospital admission. In average, there were 67 patients that fulfilled the inclusion criteria each year, with an average of 10 severely injured (ISS > 15) children annually. The quality of the collected study variables showed few missing values, with the exception of physiological values, discharge status and computer tomography (CT) findings (Additional file 1: Table S1).

This study shows that children constitute approximately 20% of all trauma patients admitted to our trauma centre. A small number of patients are characterized as severely injured, and only few patients die in the hospital following their injuries. The dominating type of injury is caused by blunt mechanism, with transport-related or falls being the most prevalent and with a tendency towards more transport-related injuries with increasing age. There is a trend towards increased use of HEMS with increasing injury severity and decreasing age. There are a low number of patients transferred from local hospitals, though almost half of these are severely injured.

Trauma is described by the World Health Organization as one of the great challenges of modern health care, representing one of the most common causes of death and disability from 1 to 44 years of age [1, 19]. There is a large variation between different parts of the world, with substantially higher rates of mortality in low-income countries [1, 20]. From 1990 to 2013, the overall consequences of road injuries among children decreased but ranked higher in 2013 (from 15 to 11) among the leading causes resulting in disability and death [20]. Our study shows that only a relatively small number of patients annually are considered potentially severely injured with an average of approximately 67 patients per year in a city population of 170,000 (Trondheim) and a regional population of 721,000 inhabitants. The annual number of severely injured children was low, meaning that very few physicians will obtain clinical experience with this patient group. Our findings are also consistent with national figures reporting a continuous decrease of fatally injured children in Norway since the 1980s [5, 21]. We found that many of the patients included were aged 15 years or older. These patients have similar physiology as adults, which limit the need for explicit paediatric expertise. Implementation of paediatric trauma teams, as seen in the USA, would therefore not seem like a cost-effective solution in Norway.

In our study, we used the ISS to define whether the patient had severe (ISS > 15), moderate (ISS 9–15) or minor (ISS 0–8) injuries. By using these definitions, only 15% of the patients included were defined as major trauma, resulting in an overtriage of 85%. Overtriage could be defined as the number of patients with minor trauma and receiving trauma team activation, whereas undertriage conversely would be severely injured patients not receiving trauma team activation [13]. The effects of overtriage would be over-utilization of costly resources, and undertriage could potentially lead to delayed treatment of patients with life-threatening injuries [13, 22]. In a study by Palmer et al., the use of a lower threshold of an ISS of 8 or higher was recommended as a better definition of a “severely injured” patient, specifically if both morbidity, mortality and the need of resources is to be assessed in paediatric trauma patients [23]. Considering this definition, a total of 33% of our patients could be defined as severely injured. We also reported NISS due to its increased predictive ability in penetrating trauma and isolated head injury and increased use in the literature [16]. In our study, a larger group of patients assessed with NISS could be defined as severely injured (NISS > 15), compared to ISS: 20% versus 15%. The clinical significance of this difference in our study is uncertain due to a low total number of severely injured patients, few deaths and short follow-up time.

We observed a total mortality rate of 1%, given that the patient is alive at hospital arrival and that a trauma team is activated. In the severely injured group, the mortality was 5%. This is lower than the mortality in our hospital among all severely injured patients, which is shown to be 12% [24]. Previous studies report that most paediatric deaths occur in the pre-hospital setting [9, 25]. Children have a different response to trauma than adults, and the injury pattern is often different [26]. We observed only small differences in physiological variables between the injury groups. This is of course limited by our large number of missing data on these variables and that normal ranges of vital parameters vary between the age groups, making it more difficult to define substantial differences. Children can compensate for injuries longer than adults compensate and sustain a normal blood pressure up until the moment they have a circulatory collapse [27]. When their vital signs ultimately are lost, attempts of rescue tend to be futile [28]. Head injuries are more common in children, due to anatomical differences. They are five times more likely to have respiratory problems caused by blunt trauma to the brain, then they are to have hypovolemic hypotension [26]. Physiological criteria for trauma team activation in children, especially circulatory variables, are therefore probably not as good an indicator of severe injury as in adults, and this age group needs a lower threshold for TTA.

The panorama of injuries in Scandinavia is different compared to many other countries, where penetrating injuries account for a much higher portion of injury mechanisms [8]. While we observed that only 2% of the patients had a penetrating trauma, studies from other parts of the world show 10 times as high percentage of penetrating trauma [28, 29]. Norway has a low incidence of gun and knife violence, and gun laws are strict. A low crime rate combined with a low availability of weapons could be the reason for penetrating trauma not being a big part of Norwegian paediatric trauma injuries [5]. Transport-related incidents are the most common cause of injuries, both globally and in Norway [1, 5]. This is also reflected within our findings, representing the patient selection of a single trauma centre. Transport-related traumas involving care were responsible for most of the minor injury patients. With new cars being safer, and having an exterior that implodes on impact to minimize forces applied to passengers, the dramatic sights on the scene of an accident may lead to a high overtriage. Globally, a marked reduction is observed in transport-related injuries [1]. The reasons for this may be due to several factors, but increased focus and use of injury prevention measures are likely to have contributed beneficially with a reduced number of injuries [5]. Although we did not observe any substantial reduction of patients throughout the study period, we observed a trend towards fewer transport-related injuries (Fig. 3), which could reflect the effects of targeted injury prevention measures. In other studies, the age group below 5 years, different major mechanisms of injury such as drowning and accidental suffocation have been reported [26]. In our study, we did not discriminate drowning or suffocation as a mechanism of injury but only registered them as “others”. It is unclear whether all near-drownings and suffocations trigger trauma team activation, and thereby would be eligible for inclusion in our study. In our patient cohort, two out of eight patients who died were reported with “others” as the mechanism of injury, and could therefore be drowning, suffocation or burn victims.

In our study, we also investigated the use of both pre- and in-hospital resources. We observed that there was a trend towards increased use of HEMS among younger age groups, also previously described [30]. In our setting, this could be explained by a lower threshold for using higher competence on the scene by bringing a specially trained anaesthesiologist to the scene, combined with a general low exposure of potentially severely injured children among EMS personnel and the need for rapid transport to a definitive care facility. Considering that our study cohort describes a 13-year time period, the use of ICU resources, performed key emergency surgical procedures and emergency procedures (e.g. intubation and thoracic drainage) remains low (Table 2). Therefore, the collective experience at this trauma centre and for each individual team member is sparse. Therefore, in order to compensate for this reduced in vivo experience of potentially severely injured children, systematic training seems warranted [31]. Similar challenges regarding low incidence and exposure to paediatric trauma are also seen in other countries [32, 33]. Introduction of targeted courses and training programmes provide greater confidence and better systematic approach to resuscitation, but the long-term effect on the outcome is still debated [32, 33]. A multiprofessional team-based approach to resuscitation of critically ill and injured children was initiated in Norway, but its effect on the outcome, system development and implementation has not yet been described [34]. Future research needs to focus on the effect of the implementation of systematic trauma care in children and its effect on the outcome.

We recognize some limitations in our study. First, due to inclusion criteria, those with severe injury and not treated by trauma teams (i.e. undertriage) were not included as no formal system within the hospital captured these patients. In the general trauma population, rates of undertriage from 10 to 19% have been described [13, 22, 35]. Second, the quality of the collected data is based on the precision of each registrar. As the components of our trauma system were developing during the study period of 13 years, several registrars were performing data sampling. Third, there was no formal regional trauma system or registry in place to ensure data capture on pre-hospital deaths; the study did not therefore include those patients who died before hospital arrival. Previous publications have described the rate of pre-hospital deaths ranging from 69 to 78% of all trauma deaths [36, 37].

In a developed healthcare system, the number of potentially severely injured children is small and with very few deaths following trauma. Transport and falls represent the most common causes of injury throughout all age groups, though with a tendency towards more transport-related injuries with increasing age. In-hospital trauma care is characterized by a low threshold for a multidisciplinary reception, low use of intensive care and need for emergency surgical procedures, though with increased need in the older children.