Introduction

Trauma continues to be a leading cause of death in patients below the age of 40 [1]. According to the WHO’s Global Status Report on Road Safety, more than 1.2 million people die on the world’s roads every year [2]. Large trauma databases document the outcome of severely injured patient. However, the majority of large databases worldwide mainly include patients admitted to a hospital. Studies focused on pre-hospital trauma deaths are rare in comparison with in-hospital observations. One reason for this may lie in difficulty to access the records of patients who died on scene or on the way to the hospital. Moreover, the documentation of these data is often insufficient to answer the research questions concerning the cause of death and the preventability of injuries incurred. In addition, the variability of inclusion criteria and definitions in publications do not allow a clear interpretation of the data. However, the analysis of causes and patterns of pre-hospital deaths is crucial to improve pre-hospital care. Current in-hospital studies describe the patterns of mortality, causes of death in severely injured patients, and they do not confirm the presence of the “trimodal” distribution of trauma deaths reported in previous decades [3, 4]. Several investigations indicate that the temporal distribution of deaths is “unimodal” or “bimodal.” Moreover, autopsy studies revealed that brain injury and severe bleeding are still the leading causes of death in trauma patients [5]. The role of these causes of deaths has not changed over the last three decades and indicates the importance of pre-hospital trauma care and trauma prevention. In this literature review, we aimed to establish the incidence of pre-hospital preventable deaths reported in studies and associated reasons for preventability, such as delay of treatment and errors in management or treatment. Moreover, we aimed to determine the current definition and conventional strategies used to define preventable deaths.

Methods

This systematic review was performed with a focus on publications with data on pre-hospital trauma preventability in severely injured patients. The PubMed/Medline database was screened for relevant publications within the last three decades (1990–2018) in the English and German languages. The following MeSH search terms were used in different combinations: “multiple trauma,” “polytrauma,” preventable deaths,” “pre-hospital deaths,” and “mortality.” Combinations and synonyms were employed for further research. Moreover, we screened references of retrieved publications in order to identify additional papers that may fit the inclusion criteria. The following inclusion criteria for publications were used: original research papers published between January 1, 1990 and January 10, 2018; articles published in peer-reviewed journals of any study design (prospective or retrospective); articles focused on pre-hospital preventable death (studies including only in-hospital data were excluded); textbook chapters, posters and abstracts; and review articles were also excluded. Publications were included if they contained required information (e.g., rate of preventable deaths and reasons for preventability).

Tables summarized the publications in a chronological manner. The following parameters were of interest: country of publication, number of patients included, preventable death rate (PP = potentially preventable and DP = definitely preventable), inclusion criteria (pre-hospital only, pre-hospital and hospital deaths), definition of preventability used in each study, type of trauma (blunt versus penetrating), study design (prospective versus retrospective), and causes for preventability mentioned within the study. Descriptive statistical analysis was performed in order to compare the data.

Results

Initially, we have performed a standardized search of literature using above-mentioned MeSH terms and identified a total of 95 publications focusing our topic. Figure 1 shows a flow diagram of studies included in the systematic review. Ninety-five papers and abstracts were screened for inclusion criteria of our systematic review. Finally, we identified 40 relevant publications, 21 of which were excluded (17 articles focused on clinical deaths only, and four articles did not include relevant data or were review papers). The remaining 19 articles (7235 death) were included in the present literature review. There were nine reports published between 1990 and 2000, 4 reports published between 2001 and 2010, and 6 articles published between 2011 and 2018.

Fig. 1
figure 1

Flow diagram of study selection according to PRISM guideline

Full size image

Table 1 describes the definition for preventability used in the included studies. The majority (12/19, 63.1%) of studies used autopsies in combination with an expert panel to assess the preventability of severe injury in the patients. Six of the nineteen (31.6%) studies used standardized scoring systems as an assessment tool, mostly in combination with other methods. In four (21%) of the studies, only one method was used. Moreover, Table 1 describes studies and their reported preventable death rate. Studies focusing only on pre-hospital trauma death reported preventable death rates from 14.6 to 47.6%. In these studies, 4.9–11.3% were DP and 25.8–42.7% were PP. In publications including both in-hospital and pre-hospital deaths, a wide range of preventable deaths (2–47.6%) were observed. Deaths were determined to be DP at 0.7–26.1% and PP at 3–37%, which corresponds to the wide range noted previously.

Table 1 Included studies with primary focus on preventable deaths after trauma
Full size table

Table 2 describes the possible reasons for preventable death found in studies. One of the most frequent (27–58%) reasons was delayed treatment of trauma victims, followed by management errors (40–60%) and treatment errors (50–76.6%). Most frequently, hemorrhagic shock management, fluid therapy, and airway control were identified as sources of treatment errors. Missed injuries were documented in up to 20% of cases.

Table 2 Summary of causes for preventability of trauma-related deaths according to included studies
Full size table

Discussion

The identification of pre-hospital preventable trauma death is used as an indicator of the efficacy and quality of a given trauma system. Upon reviewing the literature included in this study, we observed that authors used different sources to define preventable deaths. Autopsy was frequently used and is a well-established method to identify pathologic findings and other relevant information. Regardless of the advantages of this method (Table 3), autopsy studies are expensive, time consuming, and therefore not performed routinely. Several authors pointed out the importance of autopsy reports for quality control and assessment of the preventability of trauma deaths [6,7,8,9]. Another assessment tool is the preventable death panel review of deceased patients. The expert panel includes different participants in the trauma chain (e.g., anesthesiologists, emergency department doctors, trauma surgeons, neurosurgeons, radiologist, and nurses). Preventable deaths are identified by a consensus after reviewing autopsy reports, hospital records, pre-hospital information, police notes, or death certificates. Such interdisciplinary panels identify problems and may bring about corrections. However, several publications pointed out a multitude of limitations for this assessment method. Firstly, authors mentioned the importance of the panel’s composition. The lack of organization, interdisciplinarity, training, definitions used, and criteria for judgement may influence the decision in this case review process. Secondly, the value of the review panel has been questioned due to high interrater variability [10, 11]. Finally, it must be noted that an optimal trauma system and ideal circumstances (e.g., weather conditions, etc.) are assumed for these reviews of cases. Therefore, applicability of these data and conclusions may be limited.

Table 3 Description advantages and disadvantages of preventability assessments used in included studies
Full size table

The third method used for identification of preventable trauma deaths is the classification of injury severity with the Injury Severity Score (ISS), Abbreviated Injury Scale (AIS), or use of the Trauma and Injury Severity Score (TRISS) measure to determine the probability of survival [12, 13]. These standardized scores are widely used and accepted. However, more and more studies identify a low interrater reliability of ISS and AIS [14, 15]. Moreover, these scores do not include physiological parameters and might therefore be misleading. The TRISS includes physiological parameters such as blood pressure, Glasgow Coma Scale, and age. However, measuring the injury severity with the ISS or AIS comes with the same limitations.

Our systematic review of the literature revealed a wide incidence range of preventable trauma deaths reported in studies. The focus of studies with inclusion of only pre-hospital deaths still showed a high variation between studies, resulting in a rate of preventability from 14.6 to 47.6%. These data represent a relevant number of patients, which could be saved in an ideal trauma system. The large amount of variability between studies is mainly related to several points: Firstly, preventability is a difficult concept to define. Several definitions and modifications exist in the literature [2, 10, 16,17,18,19]. Most of these studies applied a combination of methods (autopsy, panel review, or scores) to identify relevant patients. Secondly, the inclusion of patients (e.g., penetrating versus blunt trauma) in each study showed large differences and variabilities. The discrepancies in study design (prospective versus retrospective) and sample size make the studies less comparable. Thirdly, other factors—such as type of trauma system, level of trauma hospital, country, geographical area or year of the analysis—may have influenced the data. Therefore, the interpretation of changes over the last decade must be performed with caution. Our analysis did not show any relevant changes in the preventability rate over recent years. While numerous studies indicate that the most preventable deaths are attributable to the hospital stay or emergency department, others pointed out that more than 50% of PP death and around 30% of DP death occurred in the pre-hospital phase in the urban trauma setting [9]. Other studies focused on the role of the injury pattern on the preventability rate. Trauma victims without traumatic brain injuries (TBIs) were associated with a higher incidence of PP and DP deaths than patients who had sustained additional brain trauma [20]. Other factors, such as car safety or alarming systems, need to be considered as well. It is still unclear whether the improvement in active or passive car safety influenced the preventable death rate. It is known that passive safety features (e.g., airbags, seat belts, whiplash protection systems, safety rules, etc.) play a role in limiting injuries caused to the drivers, passengers, and pedestrians [21,22,23,24].

The analysis of the reasons for preventable death in trauma patients revealed that delay of treatment was most frequently (27–58%) identified as a leading cause (Table 2). Severe brain injury and exsanguination were described as main causes of death in autopsy studies [5, 25, 26]. Epidemiologic parameters of pre-hospital death did not change over the last decades, according to our findings. Based on prior studies, the incidence of hemorrhagic shock (HS) as the cause of death has decreased within the last decades. Regardless, it is still a leading trauma-related cause of death after TBI. In patients with brain injury, initial airway management, early oxygenation, and resuscitation are crucial to prevent secondary brain damage [27, 28]. Decompression after intracranial bleeding is not possible prior to admission to the hospital. In patients with HS, however, early fluid therapy and bleeding control may reduce the incidence of pre-hospital death [29]. According to prior studies, the majority of avoidable deaths were in patients with treatable brain injuries combined with hypoxia [30]. Clearly, airway and hemorrhage management are key points in the pre-hospital treatment of trauma patients. The reasons for delay of treatment are multifactorial. Of course, one of the most important methods to reduce mortality is the prevention of injuries. Prevention programs—including changes in legislation and public education—have been shown to reduce incidence and mortality in road traffic injuries. Moreover, improvement in automatic systems to alarm car crashes [31, 32] allows early activation of the rescue teams and reduces the pre-hospital time and time to arrival and treatment. Depending on geographical conditions and the organization of the pre-hospital trauma system, rural parts of the country are often at a long distance from a central hospital. Several studies and databases revealed that a helicopter service might lead to a higher number of severely injured patients reaching the hospital alive [33]. The second most frequent cause of preventable trauma death identified in studies is inadequate treatment and management errors (30–81%). The majority of these errors occur after arrival in the ED. Errors in fluid therapy, blood replacement, management of thoracic injuries, and airway control were documented in studies. Standardized approaches to trauma victims (e.g., ATLS Advanced Trauma Life Support/PHTLS Prehospital Trauma Life Support) allow identification of patients at risk to die or to develop secondary injuries or complications. Education on such standardized protocols has been shown to reduce mortality and improve trauma care [34, 35]. Large numbers of medical personnel can be trained in pre-hospital management of trauma patients.

We would like to stress the point to improve the pre-hospital documentation and analysis of data. Large databases in Europe and North America deliver relevant data that allow the identification of risk factors and analysis of various management strategies. However, there is a lack of pre-hospital databases, which could shed light on pre-hospital trauma care. The majority of trauma victims die in the pre-hospital setting, and this needs to gain more attention in trauma research. We would like to encourage research groups to establish pre-hospital databases in order to address the aforementioned issues. New aims in pre-hospital research should be the implementation of databases and improved data collection for these purposes. Finally, another target for future research is the translation of novel pre-hospital interventions. More standardized techniques for pre-hospital diagnostics and hemorrhage control may be useful in the pre-hospital setting. One example is the REBOA (resuscitative endovascular balloon occlusion of the aorta) [36, 37] maneuver, which has been shown to be advantageous in severely bleeding patients. Further studies are needed to show clear indications and associated complications of this device.

Conclusion

The majority of patients die prior to hospital admission. According to our systematic review, a relevant number of patient deaths were preventable. The implementation of pre-hospital alarm systems and discussion of possible improvements and strategies for the pre-hospital trauma system may reduce the delay of treatment in the patients experiencing preventable deaths. Additionally, treatment errors could be reduced by improving education of pre-hospital services and providing standardized management. New pre-hospital interventions and techniques should be considered in order to achieve the aforementioned targets. Finally, we would like to encourage investigators to standardize their assessment techniques and documentation (e.g., implementation of databases) of pre-hospital deceased trauma victims.