Comparison of 2 thromboelastography methods using patient and control samples

Background

Viscoelastic testing at point-of-care is associated with reduced blood loss and blood product transfusions. The ROTEM (Werfen) sigma is a cartridge-based system that may facilitate point-of-care use, but limited studies exist comparing the sigma with the predicated ROTEM delta.

Objectives

We compared the performance of the ROTEM delta with that of the sigma.

Methods

Citrated blood was collected from 20 healthy donors and patients during liver transplants (n = 17), obstetrics (n = 15), cardiovascular (n = 9), and trauma surgeries (n = 10). A method comparison was performed using the delta as the predicate. Imprecision was assessed at 2 levels for each assay. Manufacturer reference intervals were verified using 20 healthy donors. An algorithm used for cardiovascular surgery with the delta was compared with the sigma.

Results

The coefficient of variation was <10% for all assays/parameters except for the thromboelastometry with extrinsic activation (EXTEM) clotting time (10.3%) and EXTEM amplitude (A)5 (10.2%). Reference intervals for the delta and sigma were comparable to manufacturer claims. The Pearson r comparing the delta and sigma exceeded .85 for all parameters/assays except for thromboelastometry with cytochalasin D-mediated platelet inhibition (FIBTEM) A10 (.77; 95% CI, .66-.86), FIBTEM A20 (.78; 95% CI, .65-.87), and thromboelastometry with heparinase clotting time (.77; 95% CI, .61-.87). No difference was observed in extrapolated thresholds from the delta-guided algorithm. However, extrapolated sigma A5 parameters for EXTEM were 5 mm lower, and for FIBTEM were 1 mm lower than delta A10 parameters.

Conclusion

The ROTEM delta and sigma devices had comparable performance. A negative bias was observed in the FIBTEM assay with lower extrapolated clinical decision points for a delta-guided treatment algorithm for the FIBTEM and EXTEM A5.