Association between thrombelastography system and thromboembolic and bleeding events in Chinese aged people

Abstract

Objectives: This study was designed to obtain the knowledge about TEG indexes distribution in Chinese aged people, as well as to test the hypothesis that previous TEG indexes are associated with the subsequent thromboembolic and bleeding events in the aged population.

Methods: We conducted a two-year follow-up study in Chinese PLA General Hospital, Beijing, China. 403 aged people were enrolled in our study. They received TEG measurements at least once when they entered this study. We collected their demographical characteristics, clinical examination information and their outcome during their observational period. Structural equation modeling (SEM) was used to analyze the relationship between the four indexes from TEG and the outcome via a pathway of indicator.

Results: We found that in the "model of bleeding" (adjusted by confounding of Anticoagulants), the model fit indices with chi-square/df = 9.555/7, CFI was 0.997, TLI was 0.994 and standardized root mean square residual (SRMR) was 0.034; while in the "model of thromboembolic events" (adjusted by confounding of Anticoagulants), the model fit indices with chi-square/df = 6.070/7, CFI was1.000, TLI was 1.002 and standardized root mean square residual (SRMR) was 0.000. The "model of thromboembolic events" showed that the four indexes (R, K, MA and ANGLE) were all significantly associated with thromboembolic events, while this significance was not found in the "model of bleeding".

Conclusions: Previous TEG indexes are significantly associated with the subsequent thromboembolic events in the aged population. Future study can test this association and provide more information for the clinical use.

Keywords: Chinese aged population; Thrombelastography (TEG); bleeding; structural equation modeling (SEM); thromboembolic event.

Figure 1

Diagrammatic representation of a TEG traces indicating the commonly reported variables.

Figure 2

The distribution of four TEG indexes in Chinese aged population.

Figure 3

Model 1 Relationship between the four indexes and bleeding. ***All paths are statistically significant (P<0.001).

Figure 4

Model 2 Relationship between four indexes and bleeding adjusted by confounding of Anticoagulants. ***All paths are statistically significant (P<0.001).

Figure 5

Model 3 Relationship between four indexes and thromboibolic events. ***All paths are statistically significant (P<0.001), **P<0.01.

Figure 6

Model 4 Relationship between four indexes and thromboibolic events adjusted by confounding of Anticoagulants. ***All paths are statistically significant (P<0.001), **P<0.01.