STRENGTHS AND LIMITATIONS OF THIS STUDY

A broad search strategy without language or geographical limitations should help ensure that all available literature is included.

Quality, risk of bias and evidence grading will be performed.

A potential limitation can be high heterogeneity in the study quality and study population. However, subgroup analysis and meta-regression will be performed to find the exact source of heterogeneity.

IntroductionBackground

Traumatic brain injury (TBI) is the leading cause of death and disability worldwide among all trauma-related injuries, with an estimated 69 million people suffering a TBI each year.1 2 Significant morbidity and mortality are commonly caused by the neurological consequences of brain injury in patients with TBI. Nevertheless, different non-neurological complications also may occur frequently following TBI and are associated with worse outcomes. These post-traumatic non-neurological complications involve the respiratory system, cardiovascular system, coagulation, liver and/or kidney.3 4

Various types of cardiovascular disturbances were found in patients with TBI, including myocardial injury, left and/or right ventricular dysfunction, electrocardiography abnormalities and cardiac arrhythmias.5 Among them, ischaemic or non-ischaemic myocardial injury, defined by cardiac troponin (cTn) elevation >99th percentile, is a relatively common cardiovascular complication, with a reported prevalence of nearly 30% in moderate to severe patients with TBI.6 However, within the current literature, there is no uniform reflection of the incidence of myocardial injury characterised by cTn elevation after TBI. The proposed mechanisms behind cTn elevation following TBI involve massive catecholamine release, overstimulation of sympathetic nerve activity and activation of the adrenal glands, resulting in myocardial injury and cardiac dysfunction.5 In recent studies, elevated cTn levels have been reported as an independent risk factor for in‐hospital mortality of TBI.7 8 In addition, Krishnamoorthy et al have found that early myocardial injury characterised by elevated high-sensitivity cTn was associated with poor 6-month clinical outcomes following moderate to severe TBI.6 Because cTn levels are widely used and the assays are relatively inexpensive, these levels are a promising marker for the risk stratification of TBI. However, many of the relevant published studies to date have varied in the severity and type of TBI among included patients as well as in their design (prospective vs retrospective), risk of bias and characteristics of patients (with cardiovascular comorbidities vs without cardiovascular comorbidities). Thus, their reported results vary considerably. Therefore, the purpose of this systematic review and meta-analysis is to estimate the cumulative incidence of early myocardial injury detected by cTn within the first 72 hours of admission among patients hospitalised after acute TBI. We also aim to synthesise the impact of myocardial injury on in-hospital mortality after acute TBI.

A previous meta-analysis has indicated that elevated cTn is significantly associated with high mortality in patients with TBI.9 However, it included only four studies with relatively small sample sizes and did not assess the incidence of elevated cTn after TBI. Furthermore, this meta-analysis did not introduce relevant analysis methods, such as subgroup analysis and meta-regression, to address the heterogeneity. Against this backdrop, we outline a protocol for conducting an updated systematic review and meta-analysis to comprehensively investigate the incidence of myocardial injury characterised by elevated cTn and the association of myocardial injury with in-hospital mortality after TBI in adults. A better understanding of the incidence and clinical impact of early myocardial injury after TBI may provide important evidence to guide future healthcare design for patients with TBI and may improve the early management of TBI.

Objectives

This study has the following objectives:

To determine the incidence of myocardial injury characterised by early cTn elevation (>99th percentile) in adult patients with acute TBI.

To identify the association between myocardial injury and in-hospital mortality in patients with acute TBI.

Methods and analysisStudy registration

The protocol is reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) statement guidelines (online supplemental appendix 1).10 This protocol was registered on the PROSPERO database (CRD42023454686) on 26 August 2023.

This systematic review was planned for April 2023 and is expected to be completed in September 2024. In the final systematic review, we will describe any deviation from the protocol.

Eligibility criteria

Studies will be selected according to the criteria outlined below.

Study designs

All cross-sectional, case-control and cohort studies will be included. Cross-sectional studies will be used to extract incidence rate data. Case-control and cohort studies will be used to extract incidence rate and/or in-hospital mortality data.

Population

The population of interest is adult patients who have sustained mild, moderate or severe TBI. Considering teenagers aged 16 and 17 years are at the legal age for consent in many countries and are often treated in adult trauma units, adults are defined as 16 years and older in our review. Due to possible variations in the diagnosis of TBI, the definitions of TBI will be extracted from source studies. If the source study definition of TBI is covered by the CDC definition,11 it will be included. Based on the initial Glasgow coma scale (GCS) score, TBI is classified as mild (GCS score, 13–15), moderate (GCS score, 9–12) and severe (GCS score, 3–8).

Exposure

The exposure will be early myocardial injury, defined as elevated serum/plasma cTn of any type (cTnI or cTnT) and assay (high-sensitive cTnI or high-sensitive cTnT) measured within the first 72 hours of admission after TBI.

According to the fourth Universal Definition of Myocardial Infarction (fourth UDMI),12 myocardial injury is defined as the presence of an elevated cTn value above the 99th percentile URL. Previous studies analysed the role of elevated cTn in TBI in general but did not differentiate between acute or chronic myocardial injury, according to the fourth UDMI. Therefore, for our study aim, the exposure will be the presence of both acute and chronic myocardial injury after TBI in our systematic review.

To detect early myocardial injury after acute TBI and improve early management of TBI, we will choose within 72 hours of admission as the time point for measuring cTn.

Outcomes

The primary outcome will be the incidence rate of myocardial injury, defined by the presence of an elevated cTn value above the 99th percentile URL within 72 hours of admission in patients with TBI. The secondary outcome will be the in-hospital mortality of patients with TBI, with or without myocardial injury.

Search strategies

A comprehensive search without language restrictions will be conducted in the Medline via Ovid, Embase via Embase.com, Cochrane Library, Scopus and Web of Science databases from inception to 1 January 2024. We will include two search terms: ‘troponin’ and ‘traumatic brain injury’. Search strategies are shown in online supplemental appendix 2. The search will be performed by two independent investigators (QR and ZL). To identify further eligible articles, references to the included studies will also be reviewed.

Inclusion criteria

Studies will be included in the review if they examine the incidence of early elevated cTn (measured within the first 72 hours of admission), the association between elevated cTn and in-hospital mortality in patients with TBI or both. In order to avoid publication bias, we will include grey literature (eg, conference abstracts, dissertations, policy documents and book chapters) and unpublished studies. Studies of mixed populations will be eligible if it is possible to extract data on patients with TBI.

Exclusion criteria

Studies will be excluded from the review if they are not original research studies with unique observational data (eg, reviews and meta-analyses). Additionally, studies will be excluded if they have a sample size smaller than 25 or if they are judged to be from the same study sample as another study included in the analysis. For studies that are identifiably from the same study sample, we will include the study with the largest sample size in the meta-analysis.

The PRISMA flow chart (see online supplemental appendix 3) will be used to present the selection process.

Data extraction

Two reviewers (JT and JH) will independently extract the relevant data from eligible studies. A data extraction spreadsheet will be used to extract data from the included studies (see online supplemental appendix 4). Consensus between the two reviewers was achieved through discussion. If no agreement can be reached, a third reviewer (XZ) will intervene. Data will be extracted from the following: study title, first author, year of publication, country of origin, study design (cohort, case control and so on), sample size, demographic data, type of cTn, outcome data and primary results for each study. We will contact study authors for clarification when the population characteristics, method of study or outcome data are unclear or not reported.

Risk of bias and quality assessment

The risk of bias will be independently assessed by two reviewers (JY and MC), with disagreements resolved by consensus and third-party (MT) adjudication if required.

The Newcastle-Ottawa Scale (NOS)13 will be used to assess the quality of case-control and cohort studies, and studies were rated as good, fair or poor (see online supplemental appendix 5). The NOS includes the following three categorical criteria with a maximum score of nine stars: ‘selection’, which accounts for a maximum of four stars; ‘comparability’, which accounts for a maximum of two stars and ‘exposure’ (case-control studies) or ‘outcome’ (cohort studies), which accounts for a maximum of three stars. The National Institutes of Health (NIH) quality assessment tool14 (see online supplemental appendix 6) will be used to assess the quality of cross-sectional studies. We will present a risk-of-bias graph and summary.

Strategy for data synthesis

Stata/MP V.14.0 (StataCorp, College Station, Texas, USA) will be used to conduct the meta-analysis. Pooled estimates of the incidence rate of elevated cTn will be determined using the ‘metaprop’ programme after Freeman-Tukey double arcsine transformation to stabilise the variances and will be calculated with 95% CIs. ORs with the corresponding 95% CIs will be used for pooled estimates of in-hospital mortality. Random-effects or fixed-effects models will be used according to the estimated heterogeneity. Heterogeneity will be investigated using the χ2 test and quantified using the I2 statistic. Low, moderate and high levels of heterogeneity are defined by I2 values of 25%, 50% and 75%, respectively. Forest plots will be used to graphically display the effect size in each study and the pooled estimates. Funnel plots and Egger’s test will be used to evaluate potential publication bias when at least 10 studies are included to synthesise the outcomes of interest.

Sensitivity analysis will be applied to identify studies responsible for heterogeneity where necessary. High heterogeneity among the available studies may be observed, including in TBI severity, presence of associated injury, prior cardiovascular comorbidity, method and timing of the cTn evaluations and management of TBI. If sufficient data are available, we will perform a comprehensive subgroup analysis based on study level (publication year, country, study design, study quality and types or assays of cTn) and patient level (associated injury, prior cardiovascular comorbidity and severity of TBI) variables to find a trend of heterogeneity. In addition, univariate and multivariate logistic meta-regression analysis will be used to find the potential source of heterogeneity. Variables with a p<0.1 in the univariate meta-regression will be included in the multivariate meta-regression. For all other analyses, a p<0.05 will be considered significant. The amount of heterogeneity in the outcome explained by risk factors will be evaluated using the R2 index.

A narrative synthesis of studies will be undertaken if the included studies are too diverse (either clinically or methodologically) to combine in a meta-analysis.

Quality of the evidence

The quality of the evidence for each outcome will be determined with the Grading of Recommendations Assessment, Development and Evaluation working group methodology in this systematic review.

Patient and public involvement

No patient is involved.

Discussion

Myocardial injury following TBI, which may negatively influence cardiac function, has been reported in observational studies. However, a robust estimate of the incidence of myocardial injury and associated in-hospital mortality after TBI is limited. With the results of our systematic review and meta-analysis, we will better understand the current situation and the unique challenges of myocardial injury after TBI, which is necessary for better management of TBI. Additionally, in our meta-analysis, identifying potential factors of heterogeneity across studies will contribute to the design of further studies.

The main strength of our study is the comprehensive analysis of outcomes and our well-recognised methodology approaches.15 16 In the first part of our work, we will examine the pooled incidence of myocardial injury characterised by elevated cTn after TBI, which has not been conducted by any other meta-analysis. Understanding the epidemiology of myocardial injury after TBI may provide useful evidence to monitor cardiac biomarkers and echocardiograms in patients with TBI for more timely and accurate treatment.5 17 In another part of our work, we will reexamine the relationship between myocardial injury and in-hospital mortality after TBI, which may provide evidence or direction for developing a prognosis model for TBI. As an update and expansion to the previous meta-analysis9, many more studies published recently will be included through our comprehensive search, and new analysis methods such as subgroup analysis and meta-regression will be introduced in our systematic review and meta-analysis.

Our study has several potential limitations. First, though we will perform our systematic review according to high methodological standards, the results of our meta-analysis may be limited by the quality of the studies and the clinical and statistical heterogeneity in the reporting of the severity of TBI, different types and assays of cTn, timing of sampling and threshold of cTn. To address such concerns, we plan to conduct subgroup analysis and meta-regressions. Second, outcomes in the studies included may not be adjusted for significant extracranial injury, a significant history of pre-existing cardiovascular disease or management of TBI, which may introduce biases. Third, our results could be biased by a small sample size due to the scarcity of studies and data on our interest outcomes.

Despite some limitations, we hope that our study can answer some important clinical questions about myocardial injury after TBI, provide useful information for clinicians, researchers and policymakers, and eventually promote better understanding and management of TBI.

Ethics statementsPatient consent for publication

Not applicable.