Viscoelastometric-guided early fibrinogen concentrate replacement during postpartum haemorrhage: OBS2, a double-blind randomized controlled trial. Collins PW et al. Br J Anaesth. 2017 Sep 1;119(3):411-421.
Postpartum haemorrhage (PPH) is the leading cause of maternal mortality worldwide. At term, the haemostatic system is prothrombotic with fibrinogen concentrations of 4–6 g litre compared with 2–4 g litre in the non-pregnant population. Fibrinogen decreases earlier than other coagulation factors during PPH suggesting a potentially important role in haemostasis. Fibrinogen <3 g litre, measured early during PPH, is associated with progression to massive haemorrhage, red blood cell (RBC) and fresh frozen plasma (FFP) transfusion, invasive procedures and level 2/3 admission. Fibrinogen >4 g litre is less often associated with progression of bleeding. During PPH, it is unknown whether fibrinogen should be maintained at a level that is normal for term (>4 g/ litre), above the non-pregnant range (>2 g /litre) or an intermediate level (3 g /litre). The clinical utility of laboratory fibrinogen to predict progression of PPH is limited because results take 60–90 min to become available. A point-of-care viscoelastometric test, Rotem® provides a Fibtem A5 result within 10 min. Fibtem correlates with laboratory fibrinogen during PPH.
The aim of this study was to identify the appropriate level of fibrinogen-related haemostasis required during PPH, by investigating whether fibrinogen replacement improved outcomes when Fibtem A5 was ≤15 mm (fibrinogen about 3 g litre).
This was a multicentre, randomized, double-blind, placebo-controlled trial, set in teaching hospital obstetric units. Women aged ≥18 yr and ≥24 weeks gestation who had ongoing major PPH (1000–1500 ml blood loss) could be enrolled. At study entry, a Fibtem was performed and samples were sent to the laboratory for full blood count, Clauss fibrinogen, prothrombin time (PT) and activated partial thromboplastin time (aPTT). Women then entered an observational phase. Fibtem was repeated after each additional 500 ml blood loss or for clinical concern. If the A5 was ≤15 mm, the baby delivered and bleeding ongoing the woman was randomized to the interventional phase. Study medication was fibrinogen concentrate or placebo. Vials contained either 1g of fibrinogen in 50ml of water or 50ml of saline. The aim was to increase Fibtem A5 to >22 mm in the fibrinogen arm. At randomization 1 g of tranexamic acid was infused. FFP was given if bleeding was ongoing or there had been an additional 500 ml of blood loss since randomization. Fibtem and coagulation screens were performed 15 min and at 24 hours after study medication.
The primary endpoint was the number of allogeneic blood products infused after study medication between the two randomized arms. Secondary endpoints were: measured blood loss, numbers of units of RBC, FFP, cryoprecipitate, and platelet transfusion within the first 24 h, cell salvage, change in Fibtem and Clauss fibrinogen, invasive procedures and initiation of breastfeeding.
Sub-group analysis of Fibtem A5 at randomization ≤12 or > 12 mm and Fibrinogen at randomization <2 or ≥ 2 g L−1 was performed.
663 women gave informed consent for study inclusion. 57/663 (8.7%)women were randomized. 55 women were analysed as 2 were found ineligible. Baseline characteristics were balanced between trial groups.
No significant difference in any of the primary or secondary study outcomes between the two groups.
Subgroup analysis showed the overall interaction between treatment arm and Fibtem A5 <12mm and fibrinogen <2 gL-1 was not significant for any outcome.
Infusion of fibrinogen concentrate triggered by Fibtem A5 ≤15 mm did not improve outcomes. Subgroup analyses suggest that fibrinogen replacement is not required if Fibtem is >12 mm or fibrinogen is >2–2.5 g litre, but an effect below those levels cannot be excluded. Studies are required that randomize women to fibrinogen replacement with Fibtem/fibrinogen <12 mm/2.5 g litre to investigate whether this is clinically significant and cost effective.
- Pertinent clinical question
- Appropriately designed trial – multicentre, randomized, double-blind, placebo-controlled trial.
- The method for randomization and blinding was clear.
- The characteristics between the two trial groups was similar.
- There was small sample numbers, despite the power analysis being based on a previous study.
- The initial fibtem and fibrinogen trigger were set too high.
- Major haemorrhages potentially missed due to recruitment issues in the emergency setting.
- Intra-uterine Deaths not included in study and other causes of PPH likely to have a consumptive coagulopathy and benefiting from early fibrinogen
- Could the combination of tranexamic acid with fibrinogen have affected the results?
This study showed no difference in outcomes when fibrinogen concentrate was infused at a fibtem <15. A larger sample size and a lower fibtem threshold for fibrinogen infusion is required to identify if there is a clinical difference.
Summary by Dr Charlotte Kingsley