Prediction of Unexplained Recurrent Miscarriages Using Thromboelastography

Objective: This study investigates the thromboelastography (TEG) changes in patients with unexplained recurrent spontaneous abortion (URSA) to identify effective diagnostic markers for URSA.

Methods: We retrospectively analyzed 160 URSA patients from the Gynecology Department of the First People's Hospital of Lianyungang (June 2017 - June 2020) and compared them with 190 healthy, fertile women without adverse pregnancy histories (control group). TEG parameters were assessed using logistic regression, applying stepwise selection for model optimization. Model performance was evaluated using Receiver Operating Characteristic (ROC) curves, determining sensitivity and specificity. The Youden index identified optimal cut points for predictive probabilities.

Results: Significant differences were observed between the URSA and control groups in coagulation reaction time (R), clot formation time (K), clot formation rate (Angle-α), and maximum clot strength (MA) (P<0.05). Multivariable logistic regression identified R, Angle-α, and MA as independent URSA risk factors. The model demonstrated excellent discrimination (AUC: 0.940; 95% CI: 0.918-0.962). The optimal cut point of predictive probability (Youden index) was P=0.355, yielding a sensitivity of 0.925 and specificity of 0.795.

Conclusion: URSA patients exhibit a hypercoagulable state even when not pregnant. More research is needed to validate our findings and explore the potential clinical implications of anticoagulants in treating URSA.