Evidence-Based Clinical Guidelines

Multidisciplinary Spine Care Antithrombotic Therapies for Adults Undergoing Spine Surgery

Second Edition

Recommendations:

Anticoagulation Continuation vs Cessation

Question 10. In patients undergoing elective surgery not performed for tumor or trauma (cervical, thoracic or lumbar), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?
Question 11. In patients undergoing spinal surgery (cervical, thoracic or lumbar) for spine trauma (with spinal cord injury and without spinal cord injury), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?
Question 12. In patients undergoing spinal surgery (cervical, thoracic or lumbar) for malignancy (with spinal cord injury and without spinal cord injury), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?

Question 10: In patients undergoing elective surgery not performed for tumor or trauma (cervical, thoracic or lumbar), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?

Continuation of Anticoagulant There is fair evidence to suggest that the use of chemoprophylaxis is not associated with increased hemorrhagic complication rates in patients undergoing elective spine surgery. Grade of Recommendation: B

Du et al1 conducted a randomized prospective trial. The purpose was the evaluation of rivaroxaban for preventing DVT after lumbar spine surgery using parnaparin as the control group. This was a prospective randomized controlled study of 665 cases who then were randomly assigned to 1 of 2 groups, 341 to the rivaroxaban (Group A) and 324 to the parnaparin group (Group B). Group A was given 10 mg orally every day for 14 days starting 6 to 8 hours after surgery. Group B was given a subcutaneous injection 40 mg every day for 14 days starting 6 to 8 hours after surgery. The results were based on efficacy endpoint indicator (all DVT, nonfatal PE, and mortality for all caused), and the safety endpoint indicator (major bleeding events). Both the incidence of thrombotic events and bleeding events, both severe and nonsevere, were found to be not significant between the 2 groups (P>0.05). It was concluded that rivaroxaban is equally as effective as parnaparin in preventing DVT and doesn’t pose any additional postoperative bleeding events. This study provides Level I evidence that rivaroxaban was equivalent to parnaparin for preventing VTE in patients undergoing lumbar spine surgery and there was no increase in bleeding adverse events. Gruber et al2 conducted a randomized controlled trial assessing bleeding complications of miniheparin (2 x 2500 IU daily)-dihydroergotamine against a placebo in 50 patients having a lumbar disc operation. Patients were randomized using a closed envelope technique and computer-produced randomization list. There was 1 (4%) DVT in the experimental group at POD5 and none in the placebo group. There were no PE noted. Patients averaged almost 11 days in the hospital postoperatively and were assessed for VTE if symptomatic. There was no increased risk of bleeding complication with the use of heparin 2500 IU sq administered twice daily. This study provides Level I evidence that the incidence of symptomatic DVT in the 10 days after lumbar disc surgery is 4%. In a prospective blinded study, Hamidi et al3 looked at the incidence of VTE and bleeding complications in patients undergoing elective instrumented spine surgery and randomly receiving either LMWH or no treatment preoperatively. All patients had postoperative compression stockings. All patients also had postoperative Doppler ultrasound. There was no significant difference in the two groups (n=89 patients) with respect to VTE incidence or hemorrhagic complications. The work group downgraded this potential Level I study due to small sample size. This paper provides Level II evidence that LMWH given preoperatively did not alter the risk of VTE or increase the risk of hemorrhagic complication. Voth et al4 published a prospective randomized double-blind investigation of 2 different regimens for VTE chemoprophylaxis. A total of 179 patients were randomized. Group I (LMWH/DHE) contained 87 patients. Each received 1500U of low molecular weight heparin with 0.5mg of dihydroergotamine once per day plus a placebo injection once per day. Group II (HDHE) contained 92 patients. Each received 5,000U sodium heparin plus 0.5mg dihydroergotamine twice per day. Both groups started treatment 2 hours before the procedure and all patients were treated for 7 days. All patients had posterior lumbar surgery for disc prolapse. All patients were hospitalized at least 7 days and received q12 hour injections. Immediately after surgery every patient received 100 microCurie of Iodine125 labelled fibrinogen. A radiofibrinogen uptake test was done daily to screen for DVT. Phlebography was done for positive screening results. There were 4/87 (4.6%) patients in LMWH/DHE group that had positive screening tests. There were 3/92 (3.3%) patients in the HDHE group that had positive screening tests. Plebography was confirmatory in 1/4 in the LMWH/DHE group and 2/3 of the HDHE group. Extent of DVT not specified. No patients had signs of PE. No patients had increased intraoperative bleeding in the LMWH/DHE but 4/92 did have increased intraoperative bleeding in HDHE group. There were 9/179 patients that received post operative transfusions with no difference between groups. There were no wound hematomas or neurologic complications related to epidural hematomas. Risk factors for DVT could not be determined for individual patients in this study given the lack of demographic and historical data on the patients. The work group downgraded this potential Level I study due to unspecified randomization method. This paper provides Level II evidence that VTE incidence after lumbar disc prolapse surgery is less than 5% when chemoprophylaxis is used without mechanoprophylaxis. The paper also provides evidence that the two regimens that were used were safe, and caused no wound hematomas or clinically significant epidural hematomas. The paper also provides evidence that there is a low incidence of intraoperative bleeding increase in one group (HDHE). Dhillon et al5 conducted a single-center, retrospective comparative study of patients who received chemoprophylaxis (1 day before and up to 3 days after surgery) compared to those who did not. The average time to initiation of chemoprophylaxis was 1.5 days after surgery. This paper provides Level III evidence that the presence of an epidural hematoma is similar with and without chemoprophylaxis. Risks of spinal epidural hematoma in patients receiving VTE prophylaxis are low. Treatment from 1-3 days after surgery is safe. Fawi et al6 published a retrospective comparative study of 2,181 patients undergoing elective thoracolumbar surgery and compared the use of anti-embolic stockings only versus subcutaneous enoxaparin given 6 hours postoperatively, using the rate of VTE, mortality, SSI, and epidural hematoma. Only symptomatic patients had either duplex ultrasound of the extremities or CT angiogram/VQ scan. In the 689 patients that received enoxaparin, there were no VTE recorded. Seven patients had a negative doppler ultrasound and 10 had a negative CT angiogram. In the 1677 patients with anti-embolic stockings only, 25 required CT angiogram, with 9 (0.5%) positives for PE, and 10 required doppler ultrasound, of which 1 was positive for DVT (0.06%). There was not an epidural hematoma identified in either group. There were more surgical site infections in the control group than the LMWH group (36 vs 5, p <0.01). This study provides Level III evidence that the incidence of DVT is <1% in elective thoracolumbar surgery using early mobilization, hydration, and anti-embolic stockings. There were no cases of epidural hematoma in patients using early mobilization, hydration, and enoxaparin at 6 hours postoperative. In a retrospective study, Fiasconaro et al7 evaluated the rate of VTE and bleeding complications after elective surgery in patients receiving chemoprophylaxis with aspirin, heparin, or LMWH. The rate of VTE was 0.13%, with no difference in incidence between the 3 treatments. The rate of hematoma was similar between the 3 treatments. The rate of blood transfusions was increased with aspirin (OR 1.48) and heparin (OR 2.01) when compared with LMWH. This paper provides Level III evidence that the rates of VTE and hematoma after surgery were similar in all 3 groups, but aspirin and heparin were associated with increased blood transfusion when compared to LMWH. Fourman et al8 conducted a retrospective comparative study assessing the use of chemophrophylaxis starting on POD 2 following elective thoracolumbar surgery, in patients deemed to be high risk for VTE. All patients received pneumatic compression devices (PCD) and either aspirin (ASA) for 30 days or fondaparinux during hospitalization and then ASA for a total of 30 days. During the study period, there were 377 identified as high risk-102 had ASA+PCD and 275 had fondaparinux+ASA+PCD. In the control group, there were 3 (2.9%) DVT and 2 (2%) PE. In the fondaparinux group, there were no DVT and 1 (0.4%) PE. There were no epidural hematoma in either group. The rates of deep infection, prolonged wound drainage, wound dehiscence, transfusion, and 90-day return to the OR were not statistically significantly different in either group. This study provides Level III evidence that the incidence of epidural hematoma, deep infection, prolonged wound drainage, wound dehiscence, transfusion, and 90-day return to the OR in elective lumbar surgery with mechanical prophylaxis and ASA is the same as the rate with this protocol and fondaparinux. There were no epidural hematoma in either group. Zeng et al9 reported the results of a historical cohort comparison between 947 patients treated with low molecular weight heparin prophylaxis compared to 814 patients operated upon previously with no chemoprophylaxis. They found a lower incidence of VTE in the treatment group compared to the historical control group (0.21% treatment group, 1.6% control group) and a slightly higher but non significant trend towards bleeding complications in the treatment group (1.8% versus 0.75%). This paper provides Level IV evidence that LMWH prophylaxis decreases VTE rate without a significant increase in bleeding complications.

References:

  1. Du W, Zhao C, Wang J, Liu J, Shen B, Zheng Y. Comparison of rivaroxaban and parnaparin for preventing venous thromboembolism after lumbar spine surgery. J Orthop Surg Res. 2015;10:78. Published 2015 May 23. doi:10.1186/s13018-015-0223-7
  2. Gruber UF, Rem J, Meisner C, Gratzl O. Prevention of thromboembolic complications with miniheparin-dihydroergotamine in patients undergoing lumbar disc operations. Eur Arch Psychiatry Neurol Sci. 1984;234(3):157-161. doi:10.1007/BF00461554
  3. Hamidi S, Riazi M. Incidence of venous thromboembolic complications in instrumental spinal surgeries with preoperative chemoprophylaxis. J Korean Neurosurg Soc. 2015;57(2):114-118. doi:10.3340/jkns.2015.57.2.114
  4. Voth D, Schwarz M, Hahn K, Dei-Anang K, al Butmeh S, Wolf H. Prevention of deep vein thrombosis in neurosurgical patients: a prospective double-blind comparison of two prophylactic regimen. Neurosurg Rev. 1992;15(4):289-294. doi:10.1007/BF00257808
  5. Dhillon ES, Khanna R, Cloney M, et al. Timing and risks of chemoprophylaxis after spinal surgery: a single-center experience with 6869 consecutive patients. J Neurosurg Spine. 2017;27(6):681-693. doi:10.3171/2017.3.SPINE161076
  6. Fawi HMT, Saba K, Cunningham A, et al. Venous thromboembolism in adult elective spinal surgery: a tertiary centre review of 2181 patients. Bone Joint J. 2017;99-B(9):1204-1209. doi:10.1302/0301-620X.99B9.BJJ-2016-1193.R2
  7. Fiasconaro M, Poeran J, Liu J, Wilson LA, Memtsoudis SG. Venous thromboembolism and prophylaxis therapy after elective spine surgery: a population-based study. Thromboembolie veineuse et traitement prophylactique après une chirurgie non urgente du rachis : une étude basée sur la population. Can J Anaesth. 2021;68(3):345-357. doi:10.1007/s12630-020-01859-2
  8. Fourman MS, Shaw JD, Nwasike CO, et al. Use of Fondaparinux Following Elective Lumbar Spine Surgery Is Associated With a Reduction in Symptomatic Venous Thromboembolism. Global Spine J. 2020;10(7):844-850. doi:10.1177/2192568219878418
  9. Zeng XJ, Peng H. Prevention of Thromboembolic Complications After Spine Surgery by the Use of Low-Molecular-Weight Heparin. World Neurosurg. 2017;104:856-862. doi:10.1016/j.wneu.2017.05.050

Cessation of Anticoagulant

A systematic review of the literature yielded no studies to adequately address this question.

Future Directions for Research

The work group has no recommendations for future research into the effect of withholding mechanical or chemical prophylaxis.

Question 11: In patients undergoing spinal surgery (cervical, thoracic or lumbar) for spine trauma (with spinal cord injury and without spinal cord injury), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?

It is suggested that the use of chemoprophylaxis is not associated with an increased hemorrhagic complication rate in patients undergoing surgery for spine trauma. Grade of Recommendation: B

Shiqing et al­1 conducted a prospective randomized study among spinal trauma patients undergoing instrumented surgery. Patients were randomized to receive low molecular weight heparin postsurgery. The incidence of DVT was significantly lower in the experimental group (5.56% vs 38.89% p=0.016). The study was classified as a Level II due to lack of blinding and small sample size. There was no increase in the readmission rate for hemorrhagic disease at one month. This study provides Level II evidence in the rate of DVT following surgical stabilization for spinal trauma. In a single-center, retrospective cohort study, Ahlquist et al2 compared patients with acute spinal cord injury who received chemoprophylaxic anticoagulation within 24 hours of surgery (early, n = 49) to those who received anticoagulation at 24-72 hours (standard, n = 20) or more than 72 hours (late, n=10). There was no increase in hematoma, seroma, or neurological deterioration after chemoprophylaxis initiation. Drain output was not increased in the early chemoprophylaxis group. This paper provides Level III evidence that early chemoprophylaxis was associated with a lower rate of DVT and without an increased risk of bleeding after surgery for SCI. In a retrospective review of a single level I trauma center, Hamidi et al3 evaluated whether administration of LMWH prior to or 48 hours following surgery had an impact on VTE occurrence and adverse events, including bleeding and intraspinal hematoma. DVT incidence was lower in the group receiving LMWH within 48 hours (2.4% vs 6.8%, P = 0.02), but no difference in PE was found (0,6% vs. 1.6%, P = 0.33). This paper provides Level III evidence that early LMWH is associated with a lower rate of DVT than late LMWH for preventing DVT in spinal trauma patients, with no difference in progression of intraspinal hematoma progression, decline in hemoglobin, or packed red blood cells transfused between the groups. In a retrospective cohort study with 2-year follow-up, Hamidi et al4 compared direct oral anticoagulants and low molecular weight heparin in an operative spinal trauma population. Direct oral anticoagulants were associated with lower rates of DVT and PE. No other differences were seen in mortality, decompression for hematoma, or transfusion of packed red blood cells. This paper provides Level III evidence that direct oral anticoagulants are associated with a lower rate of DVT and PE in an operative spinal trauma population (5.6% and 1.6%, respectively), without increasing the risk of bleeding complications. Kim et al5 conducted a retrospective cohort study at a single center over 5 years with 1432 patients. Pharmacologic prophylaxis administration prior to or 48 hours after surgery was assessed. Overall 6.2% of patients developed VTE. No complication requiring intervention was seen. This study provides Level III evidence that the use chemoprophylaxis is safe following instrumented spinal trauma surgery regardless of whether initiated before or after 48 hours from surgery. In a single-center, retrospective cohort study, Chang et al6 compared patients with acute spinal cord injury who received chemoprophylaxic anticoagulation within 48 hours to those who received anticoagulation greater than 48 hours after presentation. The rate of expansion of an intraspinal hematoma was not increased in patients who received chemoprophylaxis anticoagulation within 48 hours. This paper provides Level IV evidence that early chemoprophylaxis decreased the rate of VTE and PE risk without increasing the rate of intraspinal hematoma expansion. Sharpe et al7 studied a retrospective series among patients undergoing surgical stabilization for spinal trauma at a single institution. Patients with spinal cord trauma who received preoperative chemoprophylaxis were compared to those who received no chemoprophylaxis or less than 50% of scheduled dose. Time to surgery was 3.8 days. The overall rate of VTE was 2.8%, and was less common in patients who received chemoprophylaxis. There was no increase in postoperative hemorrhagic complications in those receiving chemoprophylaxis. This study provides Level IV evidence on the rate of VTE following surgical stabilization for spinal trauma.

References:

  1. Shiqing W, Shengzhong M, Cheng Z, Guangqing C, Chunzheng G. Efficacy of low molecular weight heparin in spinal trauma patients after part concentrated screw surgery and its influence on blood parameters and the incidence of deep venous thrombosis. Med Hypotheses. 2019;132:109330. doi:10.1016/j.mehy.2019.109330
  2. Ahlquist S, Park HY, Kelley B, Holly L, Shamie AN, Park DY. Venous Thromboembolism Chemoprophylaxis Within 24 Hours of Surgery for Spinal Cord Injury: Is It Safe and Effective?. Neurospine. 2020;17(2):407-416. doi:10.14245/ns.1938420.210
  3. Hamidi M, Asmar S, Bible L, et al. Early Thromboprophylaxis in Operative Spinal Trauma Does Not Increase Risk of Bleeding Complications. J Surg Res. 2021;258:119-124. doi:10.1016/j.jss.2020.08.029
  4. Hamidi M, Zeeshan M, Kulvatunyou N, et al. Operative spinal trauma: Thromboprophylaxis with low molecular weight heparin or a direct oral anticoagulant. J Thromb Haemost. 2019;17(6):925-933. doi:10.1111/jth.14439
  5. Kim DY, Kobayashi L, Chang D, Fortlage D, Coimbra R. Early pharmacological venous thromboembolism prophylaxis is safe after operative fixation of traumatic spine fractures. Spine (Phila Pa 1976). 2015;40(5):299-304. doi:10.1097/BRS.0000000000000754
  6. Chang R, Scerbo MH, Schmitt KM, et al. Early chemoprophylaxis is associated with decreased venous thromboembolism risk without concomitant increase in intraspinal hematoma expansion after traumatic spinal cord injury. J Trauma Acute Care Surg. 2017;83(6):1088-1094. doi:10.1097/TA.0000000000001675
  7. Sharpe JP, Gobbell WC, Carter AM, et al. Impact of venous thromboembolism chemoprophylaxis on postoperative hemorrhage following operative stabilization of spine fractures. J Trauma Acute Care Surg. 2017;83(6):1108-1113. doi:10.1097/TA.0000000000001640

Future Directions for Research

Given the relatively small number of patients in a center with spinal cord trauma, the work group recommends that future prospective study protocols perform statistical analysis to estimate the number of subjects needed to answer the primary hypothesis.

Question 12: In patients undergoing spinal surgery (cervical, thoracic or lumbar) for malignancy (with spinal cord injury and without spinal cord injury), what is the incidence of complication from continuation of anticoagulant? Cessation of anticoagulant?

A systematic review of the literature yielded no studies to adequately address this question.

Work Group Narrative

In elective spine surgery for degenerative conditions, overall incidence of VTE is very low and mechanical prophylaxis is reasonable. Patients with trauma, particularly those with SCI and patients with malignancy have higher rates of VTE and the addition of chemoprophylaxis is suggested particularly as there is no evidence to suggest increased complications from the use of chemoprophylaxis.

Future Directions for Research

Given the relatively small number of patients within a center undergoing spine surgery for malignancy, the work group recommends that future prospective study protocols perform statistical analysis to estimate the number of subjects needed to answer the primary hypothesis.