Main findings
This systematic review of contemporary RCTs recruiting multiple trauma patients found 12 different terms used to describe this population and that nearly 40% of the RCTs did not report any definition of the term used. Where definitions were included, there were more than 20 variations. These results expose that despite an international consensus in 2014 [
13], a substantial heterogeneity remains in the terms and definitions used in RCTs involving multiple trauma patients.
This diversity implies a significant between-trials heterogeneity regarding both baseline characteristics (such as the mean ISS that ranged from 11 to 32) and outcomes (such as mortality ranging between 9 to 31%). The lack of consistency in the terminology used could explain some differences in the required co-interventions, variations in the observed treatment effects and finally the low rate of significant outcomes observed in our study and in the existing literature.
Discussion with existing literature
In 2009, a systematic review including any type of studies identified 47 possible definitions of multiple trauma [
12]. Almost fifteen years later, our review underpins the persistence of this heterogeneity both in the terms used and, in the definition associated with these terms. This issue of variability in defining a specific clinical conditions for trials has been previously highlighted in various contexts, including ARDS [
65], traumatic hemorrhagic shock [
66] or refractory septic shock [
67].
The time-sensitive nature of trauma care leads to the challenge of specifying a consistent and unique definition. As an example, the selection of variables used in the definition involves a balance between the availability of variables and the timing of the intervention being evaluated. For instance, variables such as AIS or ISS are strongly associated to patient severity, making them suitable for inclusion in a definition, as proposed by the Pape et al. [
13]. However, these variables rely on the completion and interpretation of a full-body CT scan which limits the use of this definition for research projects carried out prior to scanning. Furthermore, it restricts the use of this definition in low-income settings with limited resources such as CT scans. On the other hand, some variables may be available very early after the trauma, such as blood pressure or heart rate [
3,
62], but these variables, if only considered at one time point are also likely to be less specific, potentially failing to include the population of interest.
Nevertheless, standardized consistent definitions are possible. Time-sensitive conditions have been defined as demonstrated with the Berlin definition for ARDS [
68] or SEPSIS-III definition [
69]. The strengths of these two definitions rely on objective, easily measurable and accurate clinical criteria that can be promptly measured and capture essential criteria of each syndrome. These characteristics allow for simple use and offer consistency, whether applied prospectively or retrospectively.
Limitations
First, defining the scope of a systematic review involves defining a population. This was a methodological challenge as analyzing this definition was the main aim of this systematic review. Thus, we chose to use a spectrum of synonyms of multiple trauma in the search equation and to include after titles and abstracts screening those RCTs that reported authors commitment to include trauma with a certain severity. This choice was guided by systematic reviews that have had a similar focus in other clinical situations such as polypharmacy [
70], community health workers [
71] or labor [
72]. It nevertheless leads to the exclusion of RCTs such as CRASH II that reported in their abstract the intention to include “
adult trauma patients with, or at risk of, significant bleeding” and that are not indexed in PubMed under the Mesh “
multiple trauma” [
73].
Second, due to the broad diversity of populations, interventions and outcomes included in this systematic review, it was not possible to evaluate the impact of the choice of terms or definition on the effectiveness of the interventions. This study only reports that providing a clear definition of the term used did not seems to be associated with an increased proportion of significantly positive primary outcomes.
Third, there is a possibility that some terms or definitions may have been missed, especially if the trial that used them was not within the scope of the systematic review. As a result, an additional term or definition could emerge, or another occurrence of a term or definition already included in the systematic review. Such an event would not alter the message of the review but would only emphasize the importance of the existing heterogeneity.
Fourth, this study exclusively focuses on the terms employed to denote the, though it can be argued that the true determinant of the recruited population lies within inclusion criteria. Such a statement might downplay the importance of the heterogeneity exposed in this study. Nonetheless, our findings also expose that terms such as “multiple trauma” or “severe trauma” are frequently used within the included manuscripts inclusion criteria section. Such utilization overall strengthens the problematic exposed as all these terms convey a certain degree of ambiguity. Indeed, even if widely acknowledged that, irrespective of the specific term employed, these patients are at a heightened risk of poor outcomes, an incredible diversity of potential clinical presentations exists, and this diversity can ultimately lead to the categorization of markedly distinct patients under the same generic overarching term.
Finally, it is not possible, within the context of this work, to recommend the use of one term or definition over another. The sole aim of this systematic review was to determine which terms were commonly used in the literature and which definitions were associated with these terms. Therefore, the purpose of this work was not to identify a consensus definition or to determine whether a given definition was more often associated with a significant outcome.
Implications
Also, the lack of consistency in definitions and underlying clinical heterogeneity presents a challenge for integrating previously published evidence. Meta-analysis assumes that populations are similar enough to be pooled into a single measure of effect, but this assumption is undermined when authors fail to provide a clear (or any) definition. This incomplete reporting has been shown to significantly contribute to research waste [
74].
The observed heterogeneity in definitions may also contribute to physicians' uncertainty at the bedside. In a prospective observational study, trauma surgeons have been reported to only reach a moderate agreement regarding whether a given patient should be qualified as a multiple trauma or not [
75]. This finding challenges the common belief that caregivers base their health care diagnosis on rigorous definitions and emphasize the need of a standardization of these which encompass the complexity and time-sensitive nature of trauma care.
The heterogeneity in definitions used may also reflect the presence of several phenotypes within this population, as it has been advocated for ARDS [
76]. It could indeed be argued that severe traumatic brain injury and hemorrhagic shock, as well as penetrating and non-penetrating trauma, are different diseases. In this light, it might be necessary to consensually delineate subgroups within the definition to acknowledge for these differences [
77].
Finally, for stakeholders involved in the design of future RTCs, it may be important to acknowledge that the terms used can have a direct impact on critical outcomes, such as the mortality rate. This awareness can be particularly relevant when determining the appropriate number of patients to treat.