Lloyd-Donald P(1), Zia F(2), Bellomo R(1), Hart G(1), Churilov L(3), Weinberg L(1).

Agreement and clinical interchangeability of the TEG®6s haemostatic analyser between devices, operators, and across multiple time points: a comprehensive prospective validation study.

1 Department of Intensive Care, Austin Hospital, Victoria, Australia 
2 Department of Anaesthesia, Austin Hospital, Victoria, Australia 
3The Florey Institute of Neuroscience & Mental Health, Australia

Background and Goal of Study: TEG®6s (Haemonetics Corp, USA) is a novel haemostasis analyser that measures viscoelasticity properties of blood using resonance technology. We assessed agreement of the TEG®6s analyser between devices, operators, and across multiple time points. 

Materials and Methods: We collected 3.5mL whole blood in citrated tubes from 25 adult patients in a tertiary level intensive care unit (ICU). Measurements were performed by one operator on two TEG® 6s devices (“Interdevice” agreement). Then, 5 different operators performed sample analysis from 5 healthy volunteers, using 5 TEG6S devices (“Interoperator” agreement). Finally, a single operator performed 15 measurements on 4 TEG®6s devices, with samples from 5 healthy volunteers, 5 surgical, and 5 ICU patients. Agreement across pre-set time points (0, 15, 60, 120 and 180 minutes) (“Timepoint” agreement) was examined. “Interdevice” agreement was estimated using Lin’s condordance coefficient and further validated using intraclass correlation coefficients and reduced major axis regression. “Interoperator” and “Timepoint” agreement was assessed using the Intraclass correlation coefficient estimated by a random effect regression model.

Results and Discussion: Almost perfect “Interdevice” agreement was observed, pervasive across all TEG variables. Lin’s concordance correlation coefficients (95%CI, slope, intercept) were R-time: 0.96 (0.92-0.99, 0.88, 0.57); K-time: 0.93 (0.87-0.98, 1.07, 0.00); Alpha Angle: 0.87 (0.78-0.96, 1.20, -14.10); Maximum Amplitude (MA): 0.99 (0.98-0.99, 1.02, -1.38); Clot Lysis (LY30%): 0.89 (0.82-0.97, 1.20, 0.07). “Interoperator” agreement was almost perfect for MA and LY30%, substantial for R-time, moderate for K-time, and slight for Alpha angle. The Intraclass correlation coefficients (95%CI) were R-time: 0.61 (0.21-0.99); K-time: 0.51 (0.12-0.88); Alpha Angle: 0.13 (0.00-0.91), MA: 0.84 (0.58-1.22), LY30%: 0.93 (0.78-0.99). Finally, across timepoints we observed near perfect agreement for R-time, MA and LY30%, and substantial agreement for all other variables. The Intraclass correlation coefficients (95%CI) were R-time: 0.81 (0.66-0.91); K-time: 0.67 (0.48-0.85); Alpha Angle: 0.61 (0.39-0.80), MA: 0.96 (0.91-0.98), LY30%: 0.80 (0.61-0.91). 

Conclusions: The TEG®6S platform was broadly clinically interchangeable between devices and across time points with near perfect agreement. “Interoperator” agreement was acceptable for all variables except for Alpha Angle.

Agreement and clinical interchangeability of the TEG®6s haemostatic analyser between devices, operators, and across multiple time points: a comprehensive prospective validation study