Fluid balance and outcome in critically ill patients with traumatic brain injury (CENTER-TBI and OzENTER-TBI): a prospective, multicentre, comparative effectiveness study. Wiegers EJA, Lingsma HF, Huijben JA, Cooper DJ, Citerio G, Frisvold S, Helbok R, Maas AIR, Menon DK, Moore EM, Stocchetti N, Dippel DW, Steyerberg EW, van der Jagt M. Lancet Neurol. 2021 Aug;20(8):627-638.
Patients with significant traumatic brain injury (TBI) frequently receive fluid resuscitation. However, optimal fluid management in these patients remains controversial. Large volume resuscitation can lead to worsening edema, but inadequate volume can lead to hypotension. This study used CENTER-TBI study, a cohort study across 18 European countries, and the OzENTER-TBI study from Europe and Australia, to determine the effects of intravenous fluid administration in TBI patients. Lab values, fluid administration, and outcome data were used, and the primary outcomes were ICU mortality and Glasgow Outcome Score Extended (GOSE) at 6 months from injury. Positive mean daily fluid balance was associated with higher mortality (OR 1.1) per 0.1L increase, and worse functional outcome, whereas a negative fluid balance was not associated with increased mortality or worse functional outcome.
This study attempts to address the complex topic of fluid management in TBI patients. While large and prospective, this observational study can only demonstrate association and not causation. However, it does provide compelling evidence for judicious fluid administration in the patients. The significant variation they find across centers speaks to wide differences in individual and institutional practices. This variation could be addressed with clear guidelines, especially those emphasizing close monitoring of volume status to allow for more targeted fluid administration. In the absence of such guidelines, this study suggests that resuscitation of TBI patients should target normovolemia.
Association Between Prehospital Tranexamic Acid Administration and Outcomes of Severe Traumatic Brain Injury. Bossers SM, Loer SA, Bloemers FW, Den Hartog D, Van Lieshout EMM, Hoogerwerf N, van der Naalt J, Absalom AR, Peerdeman SM, Schwarte LA, Boer C, Schober P. JAMA Neurol. 2021 Mar 1;78(3):338-345.
Tranexamic Acid (TXA) is frequently used in the treatment of severely bleeding trauma patients. In patients with severe traumatic brain injury (TBI), intracranial hemorrhage can lead to worsening brain damage. The CRASH-3 trial studied in-hospital administration of TXA in trauma patients and failed to show a benefit in severe TBI patients. The authors of this study sought to determine whether pre-hospital administration of TXA to patients with isolated severe TBI would improve neurologic outcomes or mortality. To study this, they used the data of 2,500 patients in the BRAIN-PROTECT database and compared those who received TXA to those who did not using a t-test and chi square tests and studied mortality using logistic regression. The authors found that patients who received pre-hospital TXA had higher mortality than those who did not (37% vs 30%, p=0.005). On logistic regression after adjusting for confounders, there was no difference between patients that received TXA and those that did not in the full cohort, or in those who had confirmed TBI. However, survival was significantly lower in patients with isolated severe TBI who received TXA.
This study adds to the growing body of evidence that TXA is not beneficial in severe TBI patients and may even be harmful. While this study does not examine the underlying mechanisms for their findings, the authors do discuss that TBI can be associated with coagulopathy as well is a hypercoagulable state and altering the delicate balance between these two states may be the source of the increased mortality. This study does have multiple limitations. First, it is a retrospective cohort study which can identify associations but not causal relationships. Second, the vast majority of patients received the 1g dose of TXA, so the study cannot draw any conclusions about other doses. In addition, patients in the study who received TXA had a higher injury severity score and were older, which likely also contributed to their poor outcomes. This is an intrinsic problem with pre-hospital interventions, as first responders frequently have to make complex decisions rapidly and with limited information. In the case of TBI, it is likely that other causes of depressed mental status will be confounders, such as shock or concomitant drug use.
Delayed splenic pseudoaneurysm identification with surveillance imaging. Wallen TE, Clark K, Baucom MR, Pabst R, Lemmink J, Pritts TA, Makley AT, Goodman MD. J Trauma Acute Care Surg. 2022 Jul 1;93(1):113-117.
Blunt splenic injury (BSI) affects approximately 39,000 individuals annually in the United States. As Wallen et al. highlight, while simple splenic rupture is associated with a 1% overall rate of mortality, the most feared complication of BSI – delayed splenic rupture – is associated with a 10-fold increase in the risk of mortality. Though development of pseudoaneurysm (PSA) is a well-known complication of BSI, the reported incidence of this complication has varied in the literature between 5 and 35%. As noted by the authors, there is wide variability in practice patterns with regard to the utilization of routine follow-up CT scans to assess for pseudoaneurysm.
Wallen et al. performed a retrospective analysis of 539 patients with BSI that presented to a single academic, Level I Trauma Center between 2018 and 2020 with the following injury splenic injury severities: grade I or II (n = 297, 55.1%), grade III (n = 123, 22.8%), grade IV (n = 61, 11.3%), grade V
(n = 58, 10.8%). The reported institutional protocol at the time included a plan for dedicated repeat CT-angiogram for all grade IV and V solid organ injuries as well as grade III injuries with concern for contrast blush or pseudoaneurysm on initial CT. Based on this protocol, a total of 114 patients (91% compliance rate) who initially underwent nonoperative management received follow-up imaging after a mean (SD) of 5.0 (4.4) days following admission: grade III: 87/123 (70.7%), grade IV: 22/61 (36.0%), grade V: 5/58 (8.6%)). Out of these patients, 19 were found to have splenic PSA on follow-up imaging with the following distribution: grade III 10/87 (11.5%), grade IV 7/22 (31.8%), grade V 2/5 (40%).
Based on these results, Wallen et al. continue to highlight the inherent controversy in the management of patients with BSI following initially successful nonoperative management. As noted in Aoki and Matsushima’s Letter to the Editor (published ahead of print), the use or lack thereof of prophylactic splenic angioembolization (SAE) is an important consideration when evaluating the rate of splenic pseudoaneurysm in the presented study – they argue that if patients with splenic PSA on initial CT were excluded from the present study, that the incidence of delayed PSA would approach that of the recently published SPLASH trial. However, Wallen et al. note in their reply that the reporting and success of SAE may be institution- and provider-dependent. At present, multicenter analysis may offer more generalizable insight into which patients with BSI would benefit from surveillance imaging and when.
Tranexamic acid is not inferior to placebo with respect to adverse events in suspected traumatic brain injury patients not in shock with a normal head computed tomography scan: A retrospective study of a randomized trial. Harmer JW, Dewey EN, Meier EN, Rowell SE, Schreiber MA. J Trauma Acute Care Surg. 2022 Jul 1;93(1):98-105.
The indications for and utilization of tranexamic acid in the care of trauma patients has garnered much interest and attention in recent years. From CRASH-2 to CRASH-3 to the STAAMP trial, investigators have explored who might benefit from TXA based on injury burden, presence of shock, duration of elapsed time since injury, dosage regimen, and environment (in- or out-of-hospital). For some, however, a reasonable question has remained: is there any trauma patient in 2022 who should not receive TXA?
In this retrospective analysis, the authors sought to answer this important question using prospectively collected data from the “Prehospital Tranexamic Acid Use for Traumatic Brain Injury (TXA)” trial that was also presented by Rowell et al. in “Effects of out-of-hospital tranexamic acid vs placebo on 6-month functional neurologic outcomes in patients with moderate or severe traumatic brain injury.” This randomized, multicenter trial was powered to analyze the effect of two out-of-hospital TXA dosing strategies (bolus only: 2-g out-of-hospital TXA bolus; bolus-maintenance: 1-g out-of-hospital bolus followed by 1-g infusion within two hours of injury) versus placebo on 6-month neurologic outcomes in patients age 15 or older with suspected TBI meeting the following inclusion criteria: blunt or penetrating mechanism; GCS ≤ 12 before randomization, sedatives, or paralytics; SBP ≥ 90 mmHg before randomization; IV access prior to arrival at participating trauma center. Importantly, patients were excluded if they had GCS of 3 and nonreactive pupils, ≥ 2 hours estimated since time of injury, CPR before randomization, > 20% TBSA burns, and administration of TXA or procoagulant before randomization.
In this retrospective study, however, the authors specifically focused their analysis on adverse events after receiving TXA or placebo among the 395 trial participants (41% of all enrolled participants) who did NOT have any ICH on their initial CT. Both treatment groups – bolus only and bolus-maintenance – were well-balanced compared to the placebo group with respect to age, sex, race, ethnicity, out-of-hospital GCS, injury type, ISS, and AIS Head. They efficiently summarized their own results with the following sentence: “Of the 15 adverse events individually compared in both TXA treatment groups, none were found to statistically differ from the placebo population in terms of 28-day incidence.”
The authors openly acknowledge that the original trial was only powered to assess one outcome (6-month neurological outcomes) and that their retrospective analysis is largely underpowered to assess differences in the incidence of theoretical risks of TXA including pulmonary embolism (1.5% overall), deep venous thrombosis (0.5% overall), myocardial infarction (0.7% overall), and seizures (3.0% overall). As they note in their discussion, it would likely take 6,264 individuals to power a prospective study observing the incidence of PE among patients receiving 2-g TXA versus placebo. At the end of the day, Harmer et al. face the same limitations seen in other studies that comment on adverse events following TXA administration including heterogeneous screening protocols for VTE and non-standardized characterization of seizure activity. Nonetheless, their analysis strengthens the argument that prehospital administration of TXA to patients with suspected TBI and GCS ≤ 12 should not be impeded by concern for harming those subsequently found to not have ICH on their initial CT.