Editor’s Perspective What We Already Know about This Topic Measuring the effects of factor Xa inhibitor levels to guide the management of patients needing urgent surgery or procedural interventions is not readily available. Point-of-care testing would provide clinicians with the ability to determine functional factor Xa impairment to guide management strategies. What This Article Tells Us That Is New The use of a modified thromboelastography assay demonstrated significant correlations with rivaroxaban concentrations but values were within normal ranges, and therefore clinical utility is limited. As a result, other methods to assay rivaroxaban and other Xa inhibitor concentrations are needed to determine the anticoagulant effects of these agents when needed. Background Concern remains over reliable point-of-care testing to guide reversal of rivaroxaban, a commonly used factor Xa inhibitor, in high-acuity settings. Thromboelastography (TEG), a point-of-care viscoelastic assay, may have the ability to detect the anticoagulant effect of rivaroxaban. The authors ascertained the association of apparent rivaroxaban concentration with thromboelastography reaction time, i.e. , time elapsed from blood sample placement in analyzer until beginning of clot formation, as measured using TEG and TEG6S instruments (Haemonetics Corporation, USA), hypothesizing that reaction time would correlate to degree of functional factor Xa impairment. Methods The authors prospectively performed a diagnostic accuracy study comparing coagulation assays to apparent ( i.e. , indirectly assessed) rivaroxaban concentration in trauma patients with and without preinjury rivaroxaban presenting to a single center between April 2016 and July 2018. Blood samples at admission and after reversal or 24 h postadmission underwent TEG, TEG6S, thrombin generation assay, anti–factor Xa chromogenic assay, prothrombin time (PT), and ecarin chromogenic assay testing. The authors determined correlation of kaolin TEG, TEG6S, and prothrombin time to apparent rivaroxaban concentration. Receiver operating characteristic curve compared capacity to distinguish therapeutic rivaroxaban concentration ( i.e ., greater than or equal to 50 ng/ml) from nontherapeutic concentrations. Results Eighty rivaroxaban patients were compared to 20 controls. Significant strong correlations existed between rivaroxaban concentration and TEG reaction time ( ρ = 0.67; P < 0.001), TEG6S reaction time ( ρ = 0.68; P < 0.001), and prothrombin time ( ρ = 0.73; P < 0.001), however reaction time remained within the defined normal range for the assay. Rivaroxaban concentration demonstrated strong but not significant association with coagulation assays postreversal (n = 9; TEG reaction time ρ = 0.62; P = 0.101; TEG6S reaction time ρ = 0.57; P = 0.112) and small nonsignificant association for controls (TEG reaction time: ρ = −0.04; P = 0.845; TEG6S reaction time: ρ = −0.09; P = 0.667; PT-neoplastine: ρ = 0.19; P = 0.301). Rivaroxaban concentration (area under the curve, 0.91) and TEG6S reaction time (area under the curve, 0.84) best predicted therapeutic rivaroxaban concentration and exhibited similar receiver operating characteristic curves ( P = 0.180). Conclusions Although TEG6S demonstrates significant strong correlation with rivaroxaban concentration, values within normal range limit clinical utility rendering rivaroxaban concentration the gold standard in measuring anticoagulant effect.

Background Thromboelastography is used for assessment of hemostasis. Adherence to thromboelastography-guided algorithms and aprotinin administration each decrease bleeding and blood product usage after cardiac surgery. Aprotinin, through inhibition of kallikrein, causes prolongation of the celite-activated clotting time and the activated partial thromboplastin ratio. The aim of this study was to assess the effects of aprotinin on the thromboelastography trace. Methods Three activators were used in the thromboelastography: celite (which is widely established), kaolin, and tissue factor. Assessment was performed on blood from volunteers and from patients before and after cardiac surgery. Results The tissue factor-activated thromboelastography trace was unaffected by the addition of aprotinin. When celite and kaolin were used as activators in the presence of aprotinin, the reaction time (time to clot formation) of the thromboelastography trace was prolonged (P < 0.0001) and the maximum amplitude (clot strength) was decreased (P < 0.05). With celite as an activator, the addition of aprotinin decreased (P < 0.05) the thromboelastography alpha angle (rate of clot extension). The reaction time of the celite-activated trace correlated with the activated partial thromboplastin ratio (P < 0.01). The reaction time of the tissue factor-activated trace correlated with the international normalized ratio (P < 0.01). Conclusion The thromboelastography trace is altered in the presence of aprotinin when celite and kaolin are used as activators but not when tissue factor is the activator.

Background Thrombelastograph analysis (TEG) is used to evaluate blood coagulation. Ideally, whole blood is immediately processed. If impossible, blood may be citrated and assessed after recalcification. No data describe the effect of such treatment and storage on TEG parameters. Methods Three studies were performed in 90 surgical patients. In 30 patients, blood was citrated (1:10, 0. 129 M) and recalcified (20 microl 2 M CaCl2 to 340 microl citrated blood), and TEG was performed with native blood and after recalcification after 0, 15, and 30 min of citrate storage. In another 30 patients, TEG was performed with citrated blood recalcified immediately and after 1-72 h storage. In a third study, thrombin-antithrombin complex, prothrombin fragment 1+2, and beta-thromboglobulin were measured (using enzyme-linked immunoabsorbant assay tests) at corresponding time points. Data were compared using repeated-measures analysis of variance and post hocpaired t tests. Results TEG parameters were different in recalcified citrated blood compared with native blood (P < 0.05) and changed significantly during 30-min (P < 0.025) and 72-h (P < 0.001) citrate storage. TEG parameters measured between 1 and 8 h of citrate storage were stable. Thrombin-antithrombin complex and prothrombin fragment 1+2 values were not elevated in native blood. After 30 min of citrate storage a gradual thrombin activation was observed, as evidenced by increasing thrombin-antithrombin complex and prothrombin fragment 1+2 values (P < 0.05). beta-Thromboglobulin level was increased after 2 and 8 h of citrate storage (P < 0.01). Conclusions Analysis of native blood yields the most reliable TEG results. Should immediate TEG processing not be possible, citrated blood may be used if recalcified after 1-8 h.

Background Preeclampsia is associated with a risk of abnormal hemostasis that occurs most commonly secondary to thrombocytopenia. Thromboelastography measures whole blood coagulation and has been used to manage coagulation defects in obstetric patients. The authors conducted this investigation in a large number of preeclamptic women to assess changes in coagulation using thromboelastography. Methods Thromboelastography and platelet counts were performed in 52 healthy pregnant women, 140 mild preeclamptic women, and 114 severe preeclamptic women in active labor using disposable plastic cups and pins and native whole blood. In preeclamptic patients with a platelet count <100,000/mm3, conventional coagulation tests were also performed. Epidural analgesia was provided in some women when they requested pain relief. Results Fifteen percent of all preeclamptic women (38 of 254) and 2% (1 of 52) of healthy pregnant women had a platelet count <100,000/mm3. The incidence of thrombocytopenia <100,000/mm3 was 3% (4 of 140) and 30% (34 of 114) in mild preeclamptic patients and severe preeclamptic patients, respectively. Severe preeclamptic patients with a platelet count <100,000/mm3 were significantly hypocoagulable when compared to the other study groups. Ten severe preeclamptic women with a platelet count <100,000/mm3 had a maximum amplitude <54 mm (the lower limit of maximum amplitude in healthy pregnant women enrolled in this investigation). None of the mild preeclamptic women had a maximum amplitude <54 mm. Five severe preeclamptic women with a platelet count <100,000/mm3 had an abnormal coagulation profile, whereas all four mild preeclamptic women with a platelet count <100,000/mm3 had a normal coagulation profile. Conclusion This study shows that severe preeclamptic women with a platelet count <100,000/mm3 are hypocoagulable when compared to healthy pregnant women and other preeclamptic women.