Résumé
Objective:To investigate the risk factors for bleeding and thrombosis during extracorporeal membrane oxygenation (ECMO) therapy in critically ill patients and determine the best predictors of coagulation-related complications.Methods:A retrospective analysis was performed on patients who received ECMO for respiratory or circulatory failure at Xiangya Hospital of Central South University from January 2020 to December 2022. The outcome was whether bleeding or thrombosis occurred from 24 h after ECMO insertion to before weaning. The differences in demographic characteristics, weaning conditions, prognosis, routine blood tests, organ function, coagulation and blood product transfusion of each group were compared. Logistic regression analysis was used to evaluate the risk factors for bleeding and thrombosis, and ROC curve evaluation was used to assess their capacity to predict complications.Results:A total of 61 patients with ECMO were enrolled, with 21 cases of bleeding and 14 cases of thrombosis during ECMO. Compared with the nonbleeding group, the activated partial thromboplastin time, thromboplastin time (TT), and transfusions of frozen plasma and red blood cells were higher in the bleeding group (all P<0.05). Compared with the nonthrombotic group, the increase in body weight, D-dimer (DD), fibrinogen degradation product (FDP), and improvement of arterial oxygen partial pressure (ΔPO 2) within 24 h were significantly higher in the thrombotic group (all P<0.05). Logistic regression analysis revealed that TT ( OR=1.039, 95% CI: 1.006~1.072, P=0.018) and frozen plasma transfusion volume ( OR=1.046, 95% CI: 1.010-1.083, P=0.012) were risk factors for bleeding events. FDP ( OR=1.030, 95% CI: 1.009-1.051, P=0.005), DD ( OR=1.181, 95% CI: 1.044-1.336, P=0.008), and ΔPO 2 ( OR=1.007, 95% CI: 1.002-1.012, P=0.006) were risk factors for thrombosis. According to ROC curve analysis, the AUCs of TT, frozen plasma transfusion, and combined indexes for predicting bleeding were 0.712, 0.690, and 0.816, respectively. The combined indices had a cut-off value of 0.273, a sensitivity of 75.61%, and a specificity of 80.00%. The AUCs of FDP, DD, ΔPO 2, and combined FDP with ΔPO 2 for predicting thrombosis were 0.778, 0.748, 0.786, and 0.868, respectively. The cut-off value of the combined index was 0.157, the sensitivity was 68.09%, and the specificity was 92.86%. Conclusions:TT combined with frozen plasma transfusion volume predicted bleeding optimally, while FDP plus ΔPO 2 predicted thrombotic events better during ECMO treatment in critically ill patients.
Résumé
The incidence of telbivudine-related adverse reactions has been gradually increased. The increased levels of muscle enzymes and blood lactate are common. In this case, a 23-year-old male patient with long-term oral telbivudine had a rare serious adverse reaction. The main clinical manifestations were progressive myalgia, gradually progressed to mental disorder, and together with multiple organ dysfunction, in which the level of blood lactate was increased significantly and metabolic acidosis was extremely severe. Blood purification and sodium bicarbonate were given to correct acidosis, while ceftazidime was used to prevent infection. Telbivudine was discontinued, and tenofovir disoproxil fumarate and liver protective drug were used. The patient was discharged with a better health condition. Such patients are easily misdiagnosed as neuromuscular diseases in the early stage, which might delay the treatment and worsen medical conditions. Clinicians need to be cautious and obtain an early identification to avoid misdiagnosis.