The sensitivity and specificity of rotational thromboelastometry (ROTEM) to detect coagulopathy during moderate and severe postpartum haemorrhage: a prospective observational study

Introduction: Point-of-care viscoelastic haemostatic assays such as rotational thromboelastometry (including ROTEM and TEG) have been used in the management of postpartum haemorrhage (PPH). This study compared results obtained from the automated ROTEM Sigma with laboratory tests of coagulation and platelet count during PPH. Methods: A prospective observational cohort study recruited women with PPH >= 1000 mL (or clinical concern of bleeding). The Fibtem A5, Extem CT and Pltem (Extem A5 - Fibtem A5) results were compared with laboratory tests of fibrinogen, prothrombin time (PT), activated partial thromboplastin time (APTT) and platelet count. Results: 521 women were recruited, including 274/277 (98.9%) of women with PPH >= 1500 mL. Fibtem A5 results were matched with laboratory fibrinogen in 552/644 (85.7%) samples. The incidence of abnormal laboratory results was low: fibrinogen <= 2 g/L 23/464 (5.0%), PT or APTT >1.5 x midpoint of reference range 4/464 (0.9%), and platelet count <75 x 10(9)/L 11/477 (2.3%). Area-under-the-receiver operator characteristic curve for Fibtem A5 to detect fibrinogen <= 2 g/L was 0.96 (95% Confidence Interval (CI) 0.94 to 0.98, P<0.001), with sensitivity and specificity of Fibtem A5 <= 11 mm to detect fibrinogen <= 2 g/L of 0.76 and 0.96. Prolonged Extem CT results improved after treatment of hypofibrinogenaemia alone. Intervention points for platelet and fresh frozen plasma (FFP) transfusion based on ROTEM Sigma parameters could not be established. Conclusion: During PPH (>= 1000 mL or cases of clinical concern about bleeding), ROTEM Sigma Fibtem A5 can detect fibrinogen <= 2 g/L and guide targeted fibrinogen replacement. Laboratory results should continue to be used to guide platelet and FFP transfusion.