Risk of Recurrent Venous Thromboembolism and Hemorrhage Following Interruption of Anticoagulation at Delivery

Background: Current guidelines for regional anesthesia advise for the discontinuation of prophylactic and therapeutic low molecular weight heparin (LMWH) 12 hours and 24 hours, respectively prior to the use of neuraxial anesthesia (NA) for obstetric patients. Re initiating a prophylactic dose 12 hours following epidural catheter insertion/spinal anesthesia (SA) and at least 4 hours following epidural catheter removal is also recommended. There are limited data on the recurrent risk of venous thromboembolism (VTE) and bleeding using these standards. We conducted a retrospective single center study to assess the risk of VTE and/or bleeding in pregnant women using these criteria.

Methods: Consecutive patients from 2013 to 2018 at Mount Sinai Hospital, a university affiliated tertiary care center in Toronto, Canada who were prescribed therapeutic or prophylactic LMWH antenatally and postpartum and who had NA were included. Hospital records were reviewed to determine the indication and dosage of LMWH, presence of thrombophilia, time of first LMWH injection postpartum, the mode of neuraxial anesthesia and delivery, the time of epidural catheter/spinal anesthesia, the time of epidural catheter removal, laboratory parameters and comorbid illnesses. Patients requiring therapeutic or prophylactic LMWH were assessed in the Hematology clinic and were advised to discontinue anticoagulation according to current recommendations. The primary outcomes were frequency of VTE, spinal hematoma and volume of postpartum blood loss. Statistical Analysis: Continuous variables were summarized as medians and interquartile ranges. Categorical variables were summarized as percentages. Characteristics associated with VTE and hemorrhage were analyzed using regression analysis.

Results: Of 169 pregnancies, 158 fulfilled criteria, and 110 had complete data for the time of epidural catheter removal and initiation of LMWH. Median age was 34 (IQR 5) years and median weight 90 (IQR 70) kgs. Diagnoses included antiphospholipid syndrome, Budd Chiari, provoked and unprovoked VTE. Median platelet count at delivery was 185 (IQR 76) x 10(9)/L. Thirty-three percent were using concomitant ASA antenatally. Forty-four percent (n=48) had a vaginal delivery (VD). Sixty five (59%) had epidural anesthesia, 43 (39%) had SA and, two had combined spinal/epidural anesthesia. Median time to restarting LMWH was 7.8 (IQR 4.7) hours from epidural catheter removal/spinal insertion and 9 (IQR 3.9) hours from SA. There were no spinal hematomas. Median blood loss was 500 (IQR 400) ml. One patient, who received prophylactic dose LMWH antepartum, had a Caesarean delivery, spinal anesthesia and a prophylactic dose re-started 13.7 hours after SA, developed a delayed postpartum hemorrhage. One patient developed a VTE (1%). She developed a distal and superficial thrombus immediately after delivery. She was using antenatal prophylactic LMWH as she had an unprovoked VTE predating pregnancy, was heterozygous for the prothrombin gene mutation, had epidural anesthesia and VD. She received prophylactic LMWH three hours after catheter removal which was also 10 hours after catheter insertion.

Limitations: Time of discontinuation of LMWH prior to induction could not be confirmed definitively as this was a retrospective review.

Conclusion: Prospective studies are required to confirm these findings and to determine the safety of current recommendations of interrupting anticoagulation prior to regional anesthesia and delivery and to identify risk factors for hemorrhage and recurrent VTE to optimize anticoagulation regimens for these patients.