During pregnancy, anticoagulant effect of low molecular weight heparin can be determined with both thrombelastography and anti-factor Xa activity, or not?

Venous thromboembolism is a common cause of maternal mortality. Many women need regional anaesthesia during vaginal delivery or caesarean section. However, the risk of epidural or spinal haematoma is increased if regional anaesthesia is administered during ongoing thromboprophylaxis with low molecular weight heparin (LMWH). The aims of this pilot study are to evaluate the anticoagulant effect of LMWH during pregnancy by thrombelastography and anti-factor Xa (FXa) activity and their correlations. Sixteen pregnant women were administered dalteparin or tinzaparin from early pregnancy until six weeks postpartum. The anticoagulant effect was determined immediately before and three hours after injection. The target anti-FXa activity was 0.1-0.2 IU/ml pre-injection and less than 0.45 IU/ml post-injection. Thrombelastography variables were assessed at the same time. The correlations between anti-FXa activity and thrombelastography variables were modest, with correlation coefficients at 0.4-0.5. The thrombelastography variable best correlated to anti-FXa activity <0.1 IU/ml was thrombelastography K. The area under the receiver operating characteristic curve for thrombelastography K against anti-FXa activity <0.1 IU/ml was 0.77 (95% confidence intervals 0.64 to 0.91) with sensitivity 0.76 and specificity 0.81. The correlations increased as pregnancy progressed. The anticoagulant effect of LMWH can be measured either by analysis of anti-FXa activity or with thrombelastography. However, the correlation between the two methods is too low to allow replacement of one by the other. Further studies are necessary to determine which of the methods best predicts anticoagulant effect and increased risk of bleeding complications related to delivery or regional anaesthesia.