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

Assessment of agreement and clinical interchangeability between the TEG5000® and TEG®6s thromboelastography haemostasis analysers: a 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 and TEG®5000 (Haemonetics Corp, USA) are commercial haemostasis analysers that measure viscoelasticity properties of whole blood. Both use different mechanisms to assess identical coagulation variables. The aim of this study was to assess agreement and clinical interchangeability between the TEG® 6s and TEG®5000 analysers. 

Materials and Methods: After Ethics Committee approval, we systematically collected 3.5mL whole blood in citrated tubes from 25 adult patients in a tertiary level intensive care unit (ICU). A trained operator, proficient in the use of both the TEG® 6s and TEG® 5000 systems performed all the measurements. Interdevice agreement between the TEG®6s and TEG® 5000 analysers was measured using Lin’s condordance coefficient, and further validated this using intraclass correlation coefficients and reduced major axis regression, which allowed separation of the observed bias into fixed and proportional components. 

Results and Discussion: Sixteen (64%) patients were male; the mean (SD) age was 61 (17) years. Admission ICU diagnosis: post-cardiac surgery (24%), decompensated liver disease (16%), post-liver transplantation (16%), post-general surgery (8%), neurological injury (16%), other critical illness (12%). We found that the TEG®6s and TEG® 5000 systems were broadly interchangeable. There was slight agreement, with a proportional bias of LY30% between the analysers. All other TEG variables demonstrated almost perfect or substantial agreement, with minimal fixed bias or proportional bias, with the exception of MA, which demonstrated a fixed, non-proportional bias. Between the TEG® 6s and TEG®5000 platforms, Lin’s concordance correlation coefficients (95%CI, slope, intercept) were R-time: 0.86 (0.76-0.96, 0.85, 0.41); K-time: 0.79 (0.65-0.94, 1.21, -0.87); Alpha Angle: 0.70 (0.49-0.90, 0.84, 10.74.10); Maximum Amplitude (MA): 0.90 (0.84-0.97, 0.97, -3.5); LY30%: 0.32 (0.12-0.52, 0.36, -0.03). 

Conclusions: In adult patients in the intensive care and post-operative environments, across a variety of pathophysiological states, there was almost perfect agreement in the R-time and MA, substantial agreement in K-time and Alpha Angle, and slight agreement in LY30% between the TEG® 6s and TEG®5000 analysers. With the exception of LY30%, the TEG®6s and TEG® 5000 platforms were clinically interchangeable. This has important implications for use in clinical practice and in multi-site research programs.

Assessment of agreement and clinical interchangeability between the TEG5000® and TEG®6s thromboelastography haemostasis analysers: a prospective validation study