Utility of rotational thromboelastometry in the management of massive haemorrhage at a regional Australian hospital

Rotational thromboelastometry (ROTEM) allows targeted and individualised blood product replacement.


| INTRODUCTION
Massive haemorrhage has a significant attributable morbidity and mortality and requires urgent blood product transfusion. 1,2The decision to transfuse individual blood components is often guided by clinical judgement and traditional laboratory tests, however, both have limitations in emergent situations. 3Transfusion protocols based on estimated blood loss in massive haemorrhage are not individualised and therefore do not reliably correct coagulopathy. 4Traditional coagulation tests such as platelet count, international normalised ratio (INR), activated partial thromboplastin time (aPTT), and fibrinogen level are unable to assess haemostasis as a whole and therefore may not be able to guide urgent clinical decisions. 5,6In contrast, viscoelastic testing such as rotational thromboelastometry (ROTEM Sigma, Werfen, Spain) provides prompt, quantitative data on functional coagulopathy, allowing for targeted blood product replacement.In addition, treating clinicians can perform the ROTEM testing, reducing laboratory workloads and generating results in real-time that can guide transfusion of different blood products promptly.
A recent Cochrane review found transfusion guided by viscoelastic tests reduced mortality, morbidity, and blood product utilisation. 7wever, most of the data included in this meta-analysis came from elective cardiac surgery patients at referral centres; indeed, there are few studies that compare ROTEM and non-ROTEM-guided transfusion in massive haemorrhage outside of this patient population.
][10][11] This study was therefore performed to determine if ROTEM-guided massive transfusion at a regional hospital reduced blood product utilisation and improved outcomes for patients presenting emergently with massive haemorrhage, whatever its aetiology.

| METHODS
The Far North Queensland Human Research Ethics Committee provided ethical approval for the retrospective study (EX/2022/ QCH/90519) which was conducted at Cairns Hospital, a 676-bed hospital located in the state of Queensland in tropical Australia.The hospital is the main referral centre for a population of approximately 290 000 people who live across an area of 380 000 km 2 .It has a 16-bed intensive care unit (ICU) and offers all major health specialties (medicine, surgery, women's health, paediatrics and mental health), but the hospital does not provide cardiac surgery, neurosurgery, or transplant surgery.The hospital has one ROTEM sigma machine which is located in the ICU, but all staff have access to real-time remote viewing of ROTEM results using any hospital computer.
We reviewed all cases of massive haemorrhage at the hospital from 1st July 2015 to 30th June 2017 (pre-ROTEM period) and 1st July 2019 to 30th June 2021 (post-ROTEM period).A 12-month period after the introduction of ROTEM service in December 2017 was excluded to ensure adequate hospital-wide education about ROTEM.Patients aged ≥18 years who received ≥10 units of packed red blood cells (PRBC) within 24 h or ≥four units of PRBC within 4 h for any cause of massive haemorrhage were eligible for inclusion. 8ansfusion of blood products other than PRBC prior to hospital arrival or recurrent episodes of massive transfusion during a single hospital admission were exclusion criteria.
The patients' medical records were reviewed.Data on the patients' demographics, concurrent medication use, clinical presentation, baseline laboratory findings, therapeutic interventions, and clinical course were collected.
The massive transfusion protocol (MTP) was initiated at the discretion of treating clinician.Four units of PRBC were provided in the first pack of the MTP and clinicians were encouraged to utilise ROTEM to guide transfusion.Non-ROTEM-guided transfusion was performed due to ROTEM unavailability or clinician preference.It is local policy to administer 1 g tranexamic acid (TXA) intravenously over 10 min for obstetric patients or trauma patients within 3 h of injury, followed by continuous infusion of TXA 1 g over 8 h. 12 For ROTEM-guided transfusion, the initial ROTEM was performed at MTP initiation, although transfusion was at the clinician's discretion at all times.All ROTEM tests were performed by the on-call ICU medical officer who then interpreted the results in a stepwise manner following an algorithm which was used widely in Queensland public hospitals, based on evidence collected in trauma and obstetrics patient (Appendix A). 13,14 The ICU medical officer informed the treating clinicians of the ROTEM interpretation results, and the treating clinicians were responsible for communicating with the blood bank to request the recommended blood products.Transfusion of all non-PRBC blood products transfusion was solely guided by the algorithm, with blood bank scientists only preparing the blood products when requested by the clinicians.The ROTEM test was repeated 10 min after transfusion of recommended blood products, after every four units of PRBC, or if there was ongoing bleeding.
For transfusion that was not ROTEM-guided, each subsequent pack included four units of PRBC and four units of FFP; One unit of pooled platelets was provided in every second pack starting from pack number two; 20 units of cryoprecipitate was available at the clinician's request from pack number two.
The study's primary outcome was the quantity of blood products used.Secondary outcomes included the total cost of blood products, duration of mechanical ventilation, ICU length of stay (LOS), hospital LOS, and in-hospital death.All patients with ROTEM performed were included in the intention-to-treat ROTEM-guided transfusion group.If the ROTEM algorithm was followed at all times, they were included in the per-protocol ROTEM-guided transfusion group.

| STATISTICS
Data were entered into a spreadsheet (Microsoft Excel) and analysed using statistical software (Stata 14.2).As many continuous variables had a non-parametric distribution they are presented as median with interquartile range (IQR).Groups were analysed using the Chi-square, Fisher's exact, or Kruskal-Wallis tests or logistic regression where appropriate.Outcomes were assessed using intention-to-treat and per-protocol analyses.

| RESULTS
There were 105 episodes of massive transfusion which satisfied the study's inclusion criteria, eight were excluded: in six (6%) non-PRBC blood products were received prior to hospital arrival and two (2%) were a patient's second episode of massive transfusion within the same hospital admission.This left 97 in the final analysis, 36 from the pre-ROTEM and 61 patients from the post-ROTEM period.Six episodes of massive transfusion in the post-ROTEM period had non ROTEM-guided transfusion.In all six of these cases (three ruptured aortic aneurysms and three massive upper gastrointestinal haemorrhages) the attending clinicians felt that-as the patients were exsanguinating-there was insufficient time to wait for ROTEM results before commencing transfusion.

| Intention-to-treat analysis
The baseline characteristics of two groups were similar, although there was a greater use of TXA and a higher ICU admission ratealthough a similar rate of mechanical ventilation -in the ROTEM-T A B L E 1 Characteristics, admission physiology, and baseline therapy, by statistical analyses.guided transfusion group (Table 1).The ROTEM-guided group were less likely to receive platelets and FFP but were more likely to receive fibrinogen products.The median units of PRBC transfused within 24 h was similar (Table 2).

| Per-protocol analysis
The baseline characteristics of two groups were similar.There was a higher ICU admission rate, however the rate of TXA use was similar (Table 1).The ROTEM-guided group received less PRBC, platelets, and FFP, but more fibrinogen products (Table 3).

| DISCUSSION
Meta-analyses suggest that transfusion guided by viscoelastic tests significantly reduces mortality, morbidity, and blood product use, 7 however these studies have been performed predominantly in patients having elective cardiac surgery in referral centres.There is an urgent need for studies that evaluate ROTEM-guided transfusion in other patient populations.Our study-performed in a diverse population of patients presenting acutely with massive haemorrhage to a regional Australian hospitaldemonstrated that ROTEM-guided massive transfusion was associated with a significant reduction in PRBC, FFP, and platelet utilisation.
Patients receiving ROTEM-guided massive transfusion also had a lower mortality rate than those receiving non-ROTEM-guided transfusion.
Although not statistically significant in this small sample, the reduction in mortality was similar to that reported in meta-analysis (odds ratio 0.55 in this cohort vs. 0.52 in the meta-analysis) suggesting that there may also be clinical benefit of ROTEM-guided transfusion beyond blood product conservation in this patient population.
There are significant benefits to blood product conservation in regional settings.The short shelf life of certain blood products (particularly platelets), poses a significant challenge to blood inventory management outside of tertiary settings. 15This is particularly germane in countries like Australia, where there are frequently vast distances between central and regional laboratories.There is no other source for blood products within a 400 km radius of Cairns Hospital, the setting for this study, and the delivery time for blood products ranges between 12 and 24 h.In addition, Cairns Hospital laboratory has a maximum of five pooled platelets at any time.
Although there was a significant difference in blood product utilisation, the total financial cost of blood products used were similar, as relatively expensive fibrinogen products were used more in ROTEM-guided transfusion, a function of the algorithm that is presently used in Queensland hospitals.The ROTEM algorithm that is currently employed in Cairns Hospital was devised in 2017, however, since 2018 the fibrinogen content of apheresis cryoprecipitate has increased from approximately 800 to 1012 mg/unit, an increase which is explained by the increase in the average volume of plasma collected for cryoprecipitate manufacture from 697 to 870 mL (personal communication, Dr James Daly, Australian Red Cross Lifeblood).
In step 2 of the algorithm-the fibrinogen assessment-patients can receive either cryoprecipitate or fibrinogen concentrate (Appendix A).
Therefore after 2018, the patients receiving apheresis derived cryoprecipitate in our cohort would have received more fibrinogen.Furthermore, the cost per gram of fibrinogen is likely to have been lower in patients receiving apheresis derived cryoprecipitate than those receiving whole blood derived cryoprecipitate or concentrated fibrinogen from commercial products (Appendix B).It is likely that future iterations of the ROTEM algorithm used in Queensland hospitals will take the increase in fibrinogen content of apheresis derived cryoprecipitate into consideration, and this may further reduce the costs of ROTEM-guided transfusion.
This is only the second study, to our knowledge, to examine the utility of ROTEM-guided transfusion for acute massive haemorrhage in a regional setting.Our findings were similar to those of an Indian series, which compared ROTEM-guided transfusion for acute massive haemorrhage due to trauma, surgery, postpartum haemorrhage and snake bite with standard care. 16There were 122 patients in this Indian study, 61 in each arm.Patients having ROTEM-guided transfusion had a significant reduction in PRBC, FFP, and platelet utilisation and a shorter hospital stay.The investigators reported that there was no statistical difference in mortality between the two arms, although the raw mortality data were not presented and their study, like ours, may have been underpowered to detect a difference.
Indeed, the small sample size is a significant limitation of this study as it increases the risk of type two statistical errors which may underestimate the benefit of ROTEM-guided transfusion.In this context it is notable that the scale of the salutary effects on mortality and length of stay observed in our series were similar to those reported in a large meta-analysis. 7Another limitation is the fact that ROTEM was introduced to the hospital in December 2017, later in the study period, during a time of evolving ICU capacity and growing understanding of general critical care and transfusion strategies and this may have influenced our findings.Future larger studies might examine how ROTEM, and evolution of patient management algorithms, can further refine transfusion strategies to reduce blood product use and unnecessary costs while also improving patient outcomes.It is likely that the ROTEM algorithm will evolve with greater understanding of the pathophysiology of massive haemorrhage, its response to therapeutic interventions and changes in blood product management and storage.
In conclusion, ROTEM-guided transfusion of patients presenting emergently with massive haemorrhage in this regional Australian hospital reduced the utilisation of PRBCs, FFP and platelets.These patients also had a lower overall mortality.ROTEM-guided transfusion may have particular utility in rural and remote locations where access to blood products may be limited.It also appears likely that, as in elective cardiac surgery patients, ROTEM-guided transfusion has beneficial effects on the clinical course of patients with acute haemorrhage due to other causes.