Association of thrombelastogram with postoperative hemorrhage after total joint arthroplasty by structural equation modeling
10.3969/j.issn.2095-4344.0026
- VernacularTitle:应用结构方程分析关节置换后失血程度与血栓弹力图的相关性
- Author:
Ying-Bin ZHANG
1
;
Jie XU
;
Jian-Hao WENG
;
Deng LI
;
Zhi-Qing CAI
;
Yu-Lin HUANG
;
Bao-Hua SU
;
Ruo-Fan MA
Author Information
1. 中山大学孙逸仙纪念医院关节外科
- From:
Chinese Journal of Tissue Engineering Research
2018;22(3):329-335
- CountryChina
- Language:Chinese
-
Abstract:
BACKGROUND: Postoperative anticoagulant therapy after hip and knee arthroplasties has been included in the perioperative management guidelines. However, the application of anticoagulant drugs accompanies with the risk of bleeding. Routine coagulation tests provide limited information about the quality of clots because they identify only the first stage of clotting, while thrombelastography provides a comprehensive assessment of coagulation function. But its practicality remains controversial and the research for bleeding after joint replacement is little reported. OBJECTIVE: To explore the distribution of thrombelastography parameters (time to initial fibrin formation, clotting time, α angle, and maximum amplitude) and to analyze the correlation of the four parameters with postoperative blood loss, thereby providing guidance for improving the safety and effectiveness of anticoagulant therapy. METHODS: Totally 148 patients with detection of thrombelastogram after arthroplasty from August 2015 to March 2017 in Sun Yat-sen Memorial Hospital, Sun Yat-sen University were enrolled, including 76 cases of total hip arthroplasty and 72 cases of total knee arthroplasty. Thrombelastography data were collected on day 1 postoperatively, and the perioperative blood loss was calculated. Structural equation modeling of each group was constructed to investigate the relationship of four parameters and total blood loss. RESULTS AND CONCLUSION: (1) In the structural equation modeling of hip and knee arthroplasties, the root mean square error of approximation was less than 0.08, goodness-of-fit index, adjusted goodness-of-fit index, normed fit index and comparative fit index was all higher than 0.9, and Parsi-mony goodness-of-fit index was less than 2, so the theoretical model was matched with the data. (2) There was a correlation of postoperative hemorrhage with time to initial fibrin formation, clotting time, α angle, and maximum amplitude. (3) That is to say, thrombelastogram can be used as an efficient tool in predicting bleeding after hip and knee arthroplasties. Future study based on this research will further verify the correlation and provide more information for its clinical practice.
