The main findings of this analysis are firstly, that prehospital administration of TXA in severely injured patients did not affect the incidence of myocardial injury in this cohort. Secondly, early administration of TXA is presumably not associated with the incidence of MACE in patients with severe trauma.

To date, evidence about the association of TXA administration and the incidence of myocardial injury is limited as most studies investigated the association of TXA with mortality and vasoocclusive events such as myocardial infarction.

The CRASH-2 trial investigated the effect of early administration of TXA on death, vascular occlusive events and the receipt of blood transfusion in 20,211 trauma patients [2]. In this study early administration of TXA reduced in-hospital all-cause mortality and was not associated with the incidence of myocardial infarction. Also other vascular occlusive events were not associated with the administration of TXA in this cohort. Nevertheless, the study population differed slightly from ours, as adult trauma patients with significant hemorrhage [systolic blood pressure (BP) < 90 mmHg, heart rate > 110 beats per minute or both] were included in the study. There are no data on injury severity scores. Data collection, i.e. the occurrence of myocardial infarction was recorded by an outcome form. No troponin values are reported.

Neeki et al. examined 724 patients in the Cal-PAT study and evaluated whether the administration of TXA is associated with mortality in patients with severe trauma [12]. Their findings suggest that TXA is associated with lower mortality in this cohort. There was no difference in adverse events including myocardial infarction, but they did not investigate the incidence of myocardial injury.

The findings of the STAAMP trial by Guyette et al., investigating 927 trauma patients, are as follows: administration of TXA was not associated with reduced mortality in severe trauma patients at risk for hemorrhage, which is in line with our results [13]. Further, they could not find an increased rate of vasoocclusive events, i.e. myocardial infarction, but also did not have data on troponin values and myocardial injury.

In 1310 trauma patients, prehospital TXA administration was associated with reduced mortality after 28 days [Risk ratio (RR): 0.79 (95%CI:0.63–0.99)], but not after 6 months [RR: 0.83 (95%CI: 0.67–1.03]. Moreover, it was not associated with vascular occlusive events, especially myocardial infarction [RR: 1.95 (95%CI: 0.59–6.45]. No information is given about myocardial injury [14].

In the non-cardiac surgery setting, Devereaux et al. investigated in 9,535 patients whether tranexamic acid is associated with bleeding or cardiac complications, including myocardial infarction or isolated ischemic troponin elevation. They could not show noninferiority of tranexamic acid to placebo regarding the composite cardiovascular outcome and Hazard ratio for the incidence of 30-day myocardial injury was 1.02 (95%CI: 0.91–1.14) [5].

Against this background, the results of our study can be discussed as follows: our main outcome is the incidence of myocardial injury regardless of its cause (i.e. not necessarily occlusive genesis, but also due to oxygen demand/supply imbalance), which is underrepresented in the previous studies. Nevertheless, our results match with the findings above, as TXA administration was not associated with myocardial injury in our cohort. Accumulating evidence suggests an association between myocardial injury, MACE, and mortality, in particular in non-cardiac surgery patients, but also in patients with severe trauma [15, 16]. Thus, our findings underline with special regard to myocardial injury, TXA administration seems not to be detrimental in this cohort. As our study population is defined as severely injured patients (ISS ≥16) regardless of the risk for hemorrhage, one could conclude, that the administration of TXA seems to be safe, although in patients probably not at high risk for bleeding complications.

We are aware that our study has several limitations: Firstly, TXA administration (i.e. dosing, time of administration, indication) and follow-up troponin measurements on days one and two were not according to a protocol due to the retrospective study design. This might have led to selection bias. I.e. patients with elevated troponin at presentation were more likely to receive follow-up measurements and thus patients developing myocardial injury on day 1 or 2 might be underrepresented. In the examined population, the application of tranexamic acid was carried out according to clinical indication based on the recommendations of the national guideline for the treatment of patients with polytrauma [1]. Unfortunately, there are no precise data about when exactly and in what dose the application took place. However, the hospital infrastructure around our center offers short distances and the local standard operating procedure (SOP) recommends 1 g of TXA in the preclinical setting. Therefore, one can assume that in most cases of documented administration, 1 g was administered within 3 h after trauma. We have no valid information about the incidence of severe head injuries or the incidence of hyperfibrinolysis as potential confounders after the application of TXA. Moreover, we had no information on prehospital risk factors for hemorrhage (e.g. blood pressure or heart rate) and thus could not certainly identify patients at high risk for bleeding. This could explain why only a relatively small proportion of 32% (and only 7% in prehospital setting) received TXA. These were possibly the patients in whom life-threatening bleeding was assumed. Bias was reduced by multivariable adjustment of the regression analysis and sensitivity analyses. Secondly, we could only refer to a comparatively small study group of 297 patients.