We have read with great interest the study by Baksaas-Aasen et al. [1], focusing on this clinically relevant question: in adult trauma patients presenting with signs of haemorrhagic shock, does the use of a viscoelastic haemostatic assay (VHA) augmented transfusion strategy compared to a standard strategy with conventional coagulation tests (CTT) result in an increased number of patients alive and free of massive transfusion at 24 h? The authors should be congratulated for providing such a robust clinical trial, as few randomized studies on this topic have included such a large number of patients [2]. However, we would like to discuss several points regarding this study in which no difference in overall outcome was found.

Our first concern is about the inclusion criteria. Trauma coagulopathy itself is less prevalent than expected at the time the study was designed, reflecting a likely change in trauma management practices (reduced use of crystalloids before blood products) and patient selection may have been biased as 71% of the patients received 1 g bolus of tranexamic acid prior to enrolment [3]. Less than a third of the patients presented with coagulopathy at admission (PT ratio > 1.20; Table 1). Most patients were ultimately transfused very little, since the medians of administration of plasma, fibrinogen, and platelet were null when the median (IQR) number of RBCs received by patient at enrolment was 2(1-4) (Table 2).

Second, little is reported regarding the criteria that led to the initial administration of a transfusion "pack" (RBC:plasma:platelet) before the application of the algorithms (hybrid strategy) in the two groups. This may have resulted in the administration of products that may have been unnecessary, which may have masked the expected benefit of VHA. Indeed, the proposed algorithm described in the appendix, suggests  to reach correction to the haemostatic disorders sequentially. However, it is more usual to treat haemostasis disorders simultaneously, taking into account the respective effects of the haemostasis products in order to avoid any delay, potentially fatal, in the management of severe haemorrhage.

We noticed the authors assumed that VHA guided protocol could enable a decreased in mortality or massive transfusion from 28% to 15%: it corresponds to a relative reduction of 46% in the frequency of the event, which is very ambitious, and should probably have been discussed in more details in the sample size calculation paragraph rather than in the discussion. This may also have affected the ability of the study to detect a difference. This may also have affected the ability of the study to detect a difference [4]. However, post-hoc analysis found a strong signal for the improvement in the prognosis of patients with trauma brain injury and to a lesser extent when the PT ratio was > 1.20. Moreover, study intervention was performed 21 min earlier (median) in the VHA group with more additional interventions (e.g. more fibrinogen supplementation, median dose VHA group = 4 g vs CCT = 0 g).

The ITACTIC study findings suggest above all that all-comers do not benefit from VHA-guided protocol. Future studies should focus on a better selected group of patients with significant bleeding and coagulopathy, and a high probability of administration of blood products or for specific injury (e.g. severe traumatic brain injury). They should also include a hybrid strategy with an early stepwise VHA protocol to enable a personalised and targeted transfusion strategy as early as possible [5].