Discrepancy Between Conventional Coagulation Tests and Thromboelastography During the Early Postoperative Phase of...

Background: Thromboelastography-6s (TEG^®6s), a novel device developed to assess coagulation status, presents advantages such as less frequent calibration, ease of use, and greater stability against movements compared to the previous system (TEG5000). This is the first study in the literature to compare coagulation profiles in the early postoperative period of liver resection (LR) using conventional coagulation tests (CCTs) and TEG^®6s. Methods: Forty-six adult patients admitted to the ICU post-surgery after elective LR for malignancy were included. CCTs were used to classify patients into hypocoagulable (HCG) (platelet count < 80 × 10⁹/L, international normalized ratio ≥ 1.4, or activated partial thromboplastin time > 38 s) and normocoagulable (all other cases) groups. Mann-Whitney tests, Spearman's correlation, and linear regression were used. Results: On ICU admission, nineteen (41.3%) patients had a hypocoagulable profile based on CCTs, but only two (10.5%) of them were rated as hypocoagulable by TEG (p = 0.165). Intraoperatively, HCG patients experienced higher estimated blood loss (EBL) (p = 0.002); they required more fluids (p = 0.019), and more of them received red blood cell transfusions (p = 0.025). They also had higher postoperative arterial lactate levels (p = 0.036). Postoperative 12 h EBL was similar in the two groups (around 150 mL). The ICU stay was longer for HCG group (p = 0.010). Weak associations were observed between TEG/CCTs measures of coagulation initiation [e.g., between R time citrated rapid TEG, and international normalized ratio (r² = 0.448; p < 0.001)], clot formation [i.e., between conventional fibrinogen value using Clauss method and α-angle citrated rapid TEG (r² = 0.542; p < 0.001)], and clot strength [e.g., between conventional fibrinogen and citrated kaolin maximum amplitude (r² = 0.484; p < 0.001)]. Conclusions: CCTs revealed hypocoagulability that was not confirmed by TEG^®6s. However, the thromboelastography coagulation profile was more consistent with the detected non-relevant postoperative bleeding.