SummaryIn order to compare the morphine-sparing effect, analgesic efficacy and tolerance of nefopam and propacetamol given at their highest recommended doses, 120 patients undergoing elective hepatic resection were randomly assigned to receive postoperative intravenous patient-controlled analgesia with morphine alone, or in combination with nefopam (20 mg. . Analgesia was superior in the nefopam group despite the lower morphine consumption. Adverse effects were comparable in the three groups, except for significantly more nausea in the control group (39% vs. 17 and 26% in the nefopam and propacetamol groups, respectively) and more sweating in the nefopam group (17% vs. 0 and 3% in the control and propacetamol groups, respectively). Overall patient satisfaction was better (p , 0.001) in patients given nefopam (97%) than those receiving morphine alone (82%) or propacetamol (74%).

Clinicaltrials.gov #NCT01743794.

BACKGROUND Viscoelastic techniques have made it possible to describe specific fibrinolytic phenotypes (physiological, hyperfibrinolysis and shutdown) and to establish a relationship of these phenotypes with outcome. However, there remains a debate as to whether shutdown is a state of hypercoagulability or rather a coagulopathy with moderate fibrinolysis and fibrinogen consumption. OBJECTIVES Our objectives were to describe the relationship between fibrinolytic phenotypes and outcomes, and to report the effects of tranexamic acid (TXA) administration. DESIGN This was a retrospective analysis of prospectively acquired data from a trauma registry. SETTING An academic level 1 trauma centre in the Lyon Region, from March 2011 to December 2016. PATIENTS We included all injured patients who had a rotational thromboelastometry analysis at admission. Fibrinolytic phenotypes were determined according to the maximum lysis: shutdown less than 3%, physiological 3 to 15%, hyperfibrinolysis more than 15%. MAIN OUTCOME MEASURE Mortality at 24 h and at hospital discharge. RESULTS During the study period, 473 patients were included with the following phenotypes: physiological (344 patients, 73%), shutdown (107 patients, 23%) and hyperfibrinolysis (22 patients, 5%). There was an increase in injury severity, prothrombin time ratio, fibrin degradation products and transfusion requirements from the physiological to the shutdown and hyperfibrinolysis phenotypes. Prehospital TXA administration increased the rate of shutdown and decreased the maximum lysis value at admission. After adjustment, multivariate analysis showed that fibrinolytic phenotypes, but not TXA, were independently associated with an increased risk of early death and death before hospital discharge: shutdown [odds ratio (95% confidence interval)] 2.4 (1.2 to 4.8) and hyperfibrinolysis 67.9 (7.4 to 624.2). CONCLUSION The results of the current study suggest that shutdown, which is associated with injury severity and mortality, probably reflects a moderate form of coagulopathy and fibrinolysis rather than a hypercoagulopathy. Therefore, the observation of shutdown fibrinolysis on thromboelastography/rotational thromboelastometry should not lead to withholding but rather to the administration of TXA.