Assessment of agreement between EXTEM and NATEM thromboelastometry measurement assays in critically ill neonates

Abstract

Objectives: To investigate the agreement between the EXTEM and NATEM measurements.

Methods: In this retrospective observational study, EXTEM and NATEM analyses were performed on blood samples from 162 ill neonates, providing 324 paired measurements. The agreement between EXTEM and NATEM measurements was evaluated by the nonparametric spearman's rank correlation to assess the correlation between the paired measurements, by the Bland-Altman analysis for the graphical presentation of the agreement, and by the Deming regression model to assess the significance of the agreement. The agreement between the two methods for the detection of bleeding events was determined by kappa statistic.

Results: Strong correlations were found between EXTEM and NATEM measurements for A10, MCF. The Bland-Altman plots showed good agreement for A10, MCF, LI60, and alpha angle parameters, while CT showed a nearly linear slope indicating that bias increased with the mean. The highest agreement for bleeding events was found for the A10 parameter (κ = 0.70, p < .001), while the lowest for the CT parameter (κ = 0.36, p = .94).

Conclusions: NATEM parameters that reflect clot firmness and fibrinolytic activity are strongly correlated with the corresponding EXTEM measurements with a good agreement between them, indicating that these two methods could be used interchangeably.

Keywords: EXTEM; NATEM; critically ill neonates; method comparison; rotational thromboelastometry.