Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Literature Review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Literature Review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
View/Download PDF

Translate this page into:

75 (); 212-224
doi:
10.1016/j.jor.2026.02.053

Efficacy of different types of chemoprophylaxis for venous thromboembolism in elective spine surgery: A network meta-analysis of randomized controlled trials

, , , , , , ,
a College of Medicine, Northeast Ohio Medical University, Rootstown, OH, USA
b Department of Orthopaedic Surgery, Cleveland Clinic Akron General, Akron, OH, USA
c Department of Orthopaedic Surgery, Duke University, Durham, NC, USA
d Department of Orthopaedic Surgery, University of Michigan, Ann Arbor, MI, USA

⁎Corresponding author: Nathan Cuttica. ncuttica@neomed.edu

Received: , Accepted: ,
Professor P K Surendran Memorial Education Foundation
Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Abstract

The purpose of this study was to comprehensively examine the high-level evidence regarding the efficacy and safety of chemoprophylaxis for venous thromboembolism (VTE) for adults in elective spine surgery.

A pre-registered network meta-analysis (NMA) (PROSPERO: CRD420251026334) used four databases searched through April 2025 to identify randomized controlled trials (RCTs) that examined various types of VTE chemoprophylaxis for adults in elective spine surgery. Exclusion criteria were non-randomized studies and non-elective spine surgery. This study utilized a random-effects Bayesian NMA model using risk ratio (RR) with 95% confidence intervals and surface under the cumulative ranking curve (SUCRA) for comparisons. Certainty of evidence, risk of bias, and network model transitivity were also assessed.

Twelve RCTs were included. Patients (n = 3159; mean age = 54.5 years; mean operative time = 169.0 min) were stratified into one of eight subgroups: direct oral anticoagulant (DOAC; n = 715), low-molecular-weight heparin (LMWH) (n = 814), heparin (n = 117), aspirin (n = 91), tranexamic acid (TXA; n = 363), TXA and DOAC (n = 387), warfarin (n = 35), or no chemoprophylaxis (n = 637) based on the chemoprophylaxis method received. For effectiveness, all chemoprophylactic groups demonstrated an absolute decrease in VTE and DVT incidence compared to the no chemoprophylaxis group; however, no group had a high probability (≥75%) of being the best or worst treatment, indicating no clear evidence of superiority. For safety, all subgroups demonstrated a similar incidence of significant bleeding events and epidural hematoma with no group having a high probability (≥75%) of being the best or worst treatment, indicating no clear evidence of superiority. Rates of epidural hematoma ranged from 0 to 0.56% in all studies. Certainty of evidence was “moderate” for VTE effectiveness and “low” for chemoprophylactic safety.

Various types of VTE chemoprophylaxis in elective spine surgery may be clinically equivocal for effectiveness and safety, with no method demonstrating a high probability of superiority. VTE chemoprophylaxis should be used on a case-by-case basis factoring in individual patient variables.

Keywords

Venous thromboembolism (VTE)
Elective spine surgery
Randomized controlled trials
Chemoprophylaxis
1

1 Introduction

Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is an uncommon, but serious and potentially life-threatening complication following spine surgery.1 The reported incidence of VTE after elective spine surgery is moderately low, occurring in less than 1% of patients postoperatively, according to one large cohort study.2 However, the development of VTE after spine surgery is associated with longer hospital stays, higher financial burden of healthcare, and increased morbidity.3 The most common methods to avoid VTE postoperatively include early ambulation, mechanical prophylaxis, chemoprophylaxis, or a combination of the three.4,5 Mechanical prophylaxis, such as pneumatic compression devices (PCDs) and compression stockings, are routinely used. VTE chemoprophylaxis after orthopaedic surgery is commonly in the form of low-molecular-weight heparin (LMWH) or unfractionated heparin (UFH), but there is a lack of adequate evidence supporting its use in spine surgery and a fear of adverse bleeding events, such as epidural hematoma.6

Due to an aging population and greater awareness of spinal disorders, the burden of degenerative spine disorders is increasing, thus making it imperative to mitigate postoperative complications, such as VTE.7,8 According to the most recent systematic review and meta-analysis by Muralidharan et al. (2025), routine VTE chemoprophylaxis did not demonstrate a statistically significant reduction in VTE incidence following elective spine surgery.9 However, Muralidharan et al. (2025) only conducted a pair-wise meta-analysis for patients who received VTE chemoprophylaxis versus patients who didn't receive any pharmacologic intervention. Additionally, they did not account for the relative effectiveness and safety of different classes of VTE chemoprophylaxis, hindering decision-making. Furthermore, several randomized controlled trials (RCTs) have been published since the termination of data collection performed by Muralidharan et al. (2025). While mechanical prophylaxis and early ambulation remain a mainstay in VTE prophylaxis,10 due to the multitude of chemoprophylactic agents, such as LMWH, UFH, and direct oral anticoagulants (DOACs), further research is needed to examine the efficacy in balancing the reduction of VTE and minimization of adverse bleeding events, including epidural hematomas.11

Due to the limitations of the current literature, a network meta-analysis (NMA) restricted to high-level evidence from RCTs would be an appropriate statistical technique to compare the effectiveness and safety of different anticoagulants for VTE prophylaxis in elective spine surgery. Unlike a traditional pairwise meta-analysis, NMA allows for both direct and indirect comparisons of treatment arms. Therefore, the purpose of this NMA is to enhance surgeon decision-making by comparing the effectiveness and safety of different strategies of VTE chemoprophylaxis, including no chemoprophylaxis, in elective spine surgery.

2

2 Methods

2.1

2.1 Protocol registration

This NMA compares the efficacy of different types of agents used for VTE prevention in elective spine surgery. This study was reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Network Meta-Analyses (PRISMA-NMA).12 This NMA was also pre-registered on the International Prospective Register of Systematic Reviews (PROSPERO) on April 4th, 2025, prior to study initiation (CRD420251026334). No deviations from the initial protocol were made.

2.2

2.2 Search strategy and database utilization

From database inception until April 10th, 2025, PubMed, MEDLINE, Cumulative Index to Nursing and Allied Health (CINAHL), and Web of Science were searched for articles that fit the inclusion criteria. The full search algorithm can be found in Appendix A. Also, a grey literature search of Google Scholar was performed for the first fifty results, and a full reference search of all the included articles was performed.

2.3

2.3 Inclusion and exclusion criteria

Inclusion criteria were RCTs only, that examined adults ≥18 years old that underwent elective spine surgery with at least one trial arm receiving VTE chemoprophylaxis, as well as studies published in English and Chinese. Exclusion criteria were studies that examined pediatric patients (<18 years old)13 and non-elective spinal surgery, such as those due to infection or trauma. Exclusion criteria were also studies in which at least one trial arm did not receive VTE chemoprophylaxis and studies not published in English or Chinese. Chinese articles were read by one of the co-authors.

2.4

2.4 Article search process

Following our search, articles were downloaded into Rayyan,14 duplicates were removed, articles were screened by title/abstract, and lastly screened by full-text. Article screening was performed by one author with verification of included articles completed by a second author and then finalized by the initial author.

2.5

2.5 Data extraction process

Data was extracted by one author and then independently verified for accuracy by a second author. Data extracted included first author, year of publication, study design along with inclusion and exclusion criteria, region of spine, surgery type, study grouping, patient demographics (age, gender, operative time, weight or BMI), VTE risk factors, length of hospital stay, blood loss, as well as VTE complications.

2.6

2.6 Study definitions

Study groups were divided dependent on the class of chemoprophylaxis prescribed peri- or post-operatively, regardless of dosage. Groups were split as follows: a direct oral anticoagulant (DOAC) group (rivaroxaban), low-molecular weight heparin (LMWH) group (enoxaparin, parnaparin), a heparin group (heparin-dihydroergotamine [HDHE], UFH), an antiplatelet group (aspirin), antifibrinolytic group (tranexamic acid [TXA]), a group that received TXA and DOAC, a Vitamin K antagonist group (warfarin), and a “none” group (no VTE chemoprophylaxis received). Although TXA is an antifibrinolytic and does not represent VTE chemoprophylactic medications, the subgroups containing TXA were studied simply as a comparison group as all of the included studies with TXA had at least one study arm with VTE chemoprophylactic medication to help validate our findings.

2.7

2.7 Geometry of the network and transitivity measures

The network geometry and connectivity of the NMA were analyzed, including counts of direct and indirect comparisons. To verify the appropriateness of conducting an NMA, transitivity was visually assessed across studies and groups, focusing on patient age and operative time to ensure similarity.

2.8

2.8 Within-study risk of bias or quality

All RCTs were independently assessed for bias by two authors using the Cochrane Collaboration's Tool.15 This tool examines randomization process, deviation from the intended, missing outcomes, measurement of outcomes, selection of reported results and ranks as low, high, or unclear risk of bias.

2.9

2.9 Between-study risk of bias

Due to the low number of included articles per study group, as well as small sample size of several studies included, funnel plots or Egger's test were not utilized.16 Therefore, we opted to assess publication bias by examining funding sources and study locations, as seen elsewhere in the medical literature.17

2.10

2.10 Outcome measures

Outcomes examined in the NMA included effectiveness for VTE (encompassing deep vein thrombosis [DVT] and pulmonary embolism [PE]); effectiveness of DVT as an independent endpoint; and bleeding complications, defined as author-reported severe or significant bleeding and epidural hematoma, with epidural hematoma also analyzed separately.

2.11

2.11 Certainty and inconsistency assessment

To assess for certainty of evidence, the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach was used.18 Inconsistency was evaluated for each of the NMA tests.

2.12

2.12 Statistical analysis

This NMA used the Statistical Package for the Social Sciences (version 30.0) for descriptive statistics and MetaInsight (version 6.4.0), an online statistical tool for NMA statistics.19 For the purpose of data synthesis, studies that reported values as medians [interquartile range] were converted to mean (standard deviation [SD]) via Meta-Analysis Accelerator (meta-converter.com) as supported elsewhere for comprehensiveness.20 Additionally, to further improve comprehensiveness,21 studies reporting means with missing SD were supplemented with the SD from the next closest study in terms of means, sample sizes, and/or study type. This study utilized a random-effects Bayesian NMA model using risk ratio (RR) with 95% confidence intervals [95% CI] as the effect size for binary outcomes. In cases of zero events for binary outcomes, the Bayesian model automatically continuity corrects to 0.5. Surface under the cumulative ranking curve (SUCRA) plots were generated on MetaInsight for visual representation of ranking probabilities.22 “High” probability was determined to be ≥ 75% probability, as observed elsewhere in the literature.23

3

3 Results

3.1

3.1 Search results and article grading

A total of twelve comparative studies met the inclusion criteria out of 1245 articles initially retrieved (Fig. 1)0.24–35 No articles were found via grey literature search or via a full reference search. Apart from Li et al.29 and Kashani et al.,25 there was an unclear risk of bias for all RCTs, with individual categories graded as low or unclear risk of bias with no categories being high risk (Fig. 2). Eight of the included trials reported no funding; one study had industry support, one had funding via a research grant/award, and two did not report their funding source. Five studies were conducted in Iran, four conducted in China, one in the United States of America, one in Germany, and one in Switzerland.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram.
Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram.
Risk of bias as assessed by the Cochrane Collaboration's Tool for randomized control trials.
Fig. 2 Risk of bias as assessed by the Cochrane Collaboration's Tool for randomized control trials.
3.2

3.2 Network structure and connectivity

There were a total of eighteen indirect comparisons (None versus heparin versus warfarin versus aspirin versus TXA + DOAC; LMWH versus warfarin versus aspirin versus TXA versus TXA + DOAC; heparin versus warfarin versus aspirin versus TXA versus TXA + DOAC; DOAC versus TXA; warfarin versus aspirin versus TXA versus TXA + DOAC; aspirin versus TXA versus TXA + DOAC) and ten possible direct comparisons (None versus LMWH; None versus TXA; None versus DOAC; LMWH versus DOAC; LMWH versus heparin; DOAC versus heparin; DOAC versus warfarin; DOAC versus aspirin; TXA versus TXA + DOAC; TXA + DOAC versus DOAC) for total VTE events (Fig. 3). Eighteen indirect and eleven direct comparisons were made for DVT occurrence (Fig. 4). Twelve indirect and nine direct comparisons were made for incidence of adverse bleeding events (Fig. 5). Four indirect and six direct comparisons were made for incidence of epidural hematoma (Fig. 6).

Network plot of all studies examining venous thromboembolism (VTE). The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Fig. 3 Network plot of all studies examining venous thromboembolism (VTE). The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Network plot of all studies examining deep vein thrombosis (DVT). The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Fig. 4 Network plot of all studies examining deep vein thrombosis (DVT). The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Network plot of all studies examining adverse bleeding events. The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Fig. 5 Network plot of all studies examining adverse bleeding events. The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Network plot of all studies examining epidural hematoma. The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
Fig. 6 Network plot of all studies examining epidural hematoma. The size of nodes and thickness of edges represent the number of studies that examined a treatment and compared two given treatments, respectively. Abbreviations: LMWH, low-molecular-weight heparin; DOAC, direct oral anticoagulant; TXA, tranexamic acid.
3.3

3.3 Patient demographics

Patient demographics, measured outcomes, and interventions of the studies included are shown in Table 1. Patients (n = 3159) had a frequency weighted mean (FWM) age (SD) of 54.5 (12.9) years and FWM operative time of 169.0 (87.3) minutes. Du et al. (2015)24 did not report mean age of participants and Du et al. (2015),24 Gruber et al. (1984),25 Hamidi et al. (2015),26 Nikouei et al. (2022),30 and Yang et al. (2017)34 did not report mean operative time. All participants (n = 3159) were stratified into one of eight groups: DOAC group (n = 715), LMWH group (n = 814), heparin group (n = 117), aspirin group (n = 91), tranexamic acid (TXA) group (n = 363), TXA + DOAC group (n = 387), warfarin group (n = 35), or no chemoprophylaxis (None) group (n = 637). Sufficient between-study transitivity based on mean age was recorded (Fig. 7). Between-study transitivity based on mean operative time was also recorded and was generally acceptable. By group, patients in the DOAC group (n = 715) had a FWM age (SD) of 52.5 (13.6) years with a mean operative time of 164.7 (54.5) minutes. Patients in the LMWH group (n = 814) had a FWM age of 52.3 (15.2) years and FWM operative time of 146.7 (80.9) minutes. The heparin group (n = 117) had a FWM age of 51.6 (10.3) years and FWM operative time of 60.8 (22.3) minutes. The aspirin group (n = 91) had a FWM age of 52.9 (15.2) and FWM operative time of 254.4 (53.4) minutes. The group that received TXA (n = 363) had a FWM age of 55.2 (10.3) years and FWM operative time of 181.2 (52.5) minutes. The group that received TXA and a DOAC (TXA + DOAC; n = 387) had a FWM age of 57.5 (10.5) years and a FWM operative time of 178.8 (62.5) minutes. The warfarin group (n = 35) had a FWM age of 46.0 (13.8) years and FWM operative time of 232.0 (89.0) minutes. Lastly, the group that received no chemoprophylaxis (None; n = 637) had a FWM age of 54.8 (13.4) years and FWM operative time of 196.3 (79.6) minutes. Six RCTs included only evaluated lumbar spine surgery, comprising 70.7% of our NMA population. The remaining six RCTs either included various regions of the spine or regions were not specified by the authors.

Table 1 Baseline characteristics of included studies. DVT, deep vein thrombosis; PE, pulmonary embolism; PO, per oral; SQ, subcutaneous; DHE, dihydroergotamine; LMWH, low-molecular-weight heparin; VTE, venous thromboembolism; TXA, tranexamic acid; TED, thromboembolic deterrent; SCD, sequential pneumatic compression devices; UFH, unfractionated heparin. Hyphen represents that the study did not report or record the associated outcome.
First author, year Sample Size Average age (years) % Female Measured Outcomes Follow-up Treatment Groups/Administration Details
Du, 2015 341 - - DVT, PE, bleeding - 10 mg rivaroxaban PO initiated 6-8 h post-operative daily for up to 14 days
324 - - 40 mg parnaparin SQ injection initiated 6-8 h post-operative daily for up to 14 days
Gruber, 1984 25 47.4 ± 13 36 DVT, PE, bleeding - 2500 IU heparin-DHE SQ injection initiated 2 h pre-operative then twice daily for 7 days or until discharge
25 44.5 ± 9.7 28 Placebo SQ injection initiated 2 h pre-operative then twice daily for 7 days or until discharge
Hamidi, 2015 40 53.4 ± 15.7 53 DVT, PE, bleeding 8 months 40 mg LMWH SQ injection within 12 h pre-operative then once daily until discharge
49 50.1 ± 13.8 55 Postoperative compression stockings
Kashani, 2025 110 53.8 ± 13.8 52 DVT, PE - 10 mg rivaroxaban PO tablet until ambulatory
110 51.2 ± 14.2 55 1 mg/kg/day enoxaparin SQ injection until ambulatory
Kavian, 2024 50 44.5 ± 14.5 28 DVT, bleeding 7 days 81 mg aspirin PO daily tablet initiated 24 h post-operative and continued for 7 days
50 39.5 ± 18.0 36 40 mg enoxaparin SQ injection initiated 24 h post-operative and continued for 7 days
Li, 2023 218 56.8 ± 10.5 64 VTE, bleeding 5 days 2 g TXA IV injection and wound soaked in TXA solution prior to incision closure and 10 mg rivaroxaban PO daily tablet for 35 days
212 55.5 ± 10.6 65 2 g TXA IV injection and wound soaked in TXA solution prior to incision closure
227 55.3 ± 10.4 63 0.9% NaCl IV 15 min prior to skin incision
Nikouei, 2022 41 63.2 ± 7.1 61 DVT 3 months 325 mg aspirin PO initiated post-operative day one and continued daily for 12 weeks
41 64.3 ± 6.6 66 No pharmacologic intervention
Rokito, 1996 42 44 ± 11.4 60 DVT, PE, bleeding >1 year Thigh-high TED compression stockings
33 45 ± 14.2 61 Thigh-high TED compression stockings and SCD wraps to the calf and thighs
35 46 ± 13.8 60 Thigh-high TED compression stockings and 10 mg coumadin PO loading dose initiated evening before surgery then daily adjusted doses to maintain prothrombotic time 1.3-1.5 times the control for 5-7 days
Shafiei, 2022 123 51.8 ± 13.0 37 VTE, bleeding 25.8 months 10 mg rivaroxaban PO tablet initiated post-operative day one and continued up to 14 days after discharge
121 52.4 ± 12.2 40 1 mg/kg/day enoxaparin SQ injection initiated post-operative day one and continued up to 14 days after discharge
Voth, 1992 87 52.5 ± 8.2 56 DVT, bleeding 8 days 32 mg LMWH with 0.5 mg DHE daily and placebo injection daily initiated 2 h pre-operative for 7 days
92 52.8 ± 9.1 55 5000 U UFH and 0.5 mg DHE twice daily initiated 2 h pre-operative for 7 days
Yang, 2017 82 61 ± 17.7 51 DVT, bleeding - 0.4 mL (4100 IU) LMWH SQ injection initiated 12 h post-operative once daily for 7-14 days
82 61 ± 19.1 50 Control group received no intervention
Zhang, 2020 138 57.0 ± 10.2 66 VTE, bleeding 35 weeks 100 mL 0.9% NaCl 15 min before skin incision
151 54.7 ± 9.9 62 100 mL 0.9% NaCl 15 min before skin incision, 1 g TXA IV, and wound soaked in TXA solution prior to incision closure
141 59.2 ± 10.2 60 10 mg rivaroxaban PO daily for 35 days post-operative
169 58.3 ± 10.5 69 100 mL 0.9% NaCl 15 min before skin incision, 1 g TXA IV, wound soaked in TXA solution prior to incision closure, and 10 mg rivaroxaban PO daily for 35 days post-operative
Measure of between-study transitivity by mean age (top) and mean operative time (bottom). Abbreviations: H, heparin; LMWH, low-molecular weight heparin; DOAC, direct oral anticoagulant; ASA, aspirin; TXA, tranexamic acid; T + D, TXA and DOAC; W, warfarin. Error bars represent standard deviation (SD).
Fig. 7 Measure of between-study transitivity by mean age (top) and mean operative time (bottom). Abbreviations: H, heparin; LMWH, low-molecular weight heparin; DOAC, direct oral anticoagulant; ASA, aspirin; TXA, tranexamic acid; T + D, TXA and DOAC; W, warfarin. Error bars represent standard deviation (SD).
3.4

3.4 Incidence of VTE by anticoagulant administered

Incidence of VTE, DVT, bleeding complications, and epidural hematoma are demonstrated in Table 2.

Table 2 Occurrence of venous thromboembolism (VTE), deep vein thrombosis (DVT), bleeding complications, and epidural hematoma (EDH) per study. Hyphen represents that the study did not report or record the associated outcome. Abbreviations: PE, pulmonary embolism.
First Author, Year Sample Size Group VTE: DVT + PE (%) DVT (%) Bleeding Complications (%) EDH (%)
Du, 2015 341 DOAC 7 (2.1) 6 (1.8) 24 (7.0) 1 (0.3)
324 LMWH 12 (3.7) 10 (3.1) 18 (5.6) 0 (0.0)
Gruber, 1984 25 Heparin 1 (4.0) 1 (4.0) 7 (28.0%) -
25 None 0 (0.0) 0 (0.0) 9 (36%) -
Hamidi, 2015 40 LMWH 1 (2.5) 0 (0.0) - -
49 None 2 (4.1) 2 (4.1) - -
Kashani, 2025 110 DOAC 4 (3.6) 4 (3.6) 0 (0.0) 0 (0.0)
110 LMWH 4 (3.6) 2 (1.8) 0 (0.0) 0 (0.0)
Kavian, 2024 50 Aspirin 2 (4.0) 2 (4.0) 2 (4.0) -
50 LMWH 5 (10.0) 5 (10.0) 1 (2.0) -
Li, 2023 218 TXA + DOAC 8 (3.7) 8 (3.7) 0 (0.0) 0 (0.0)
212 TXA 34 (16.0) 34 (16.0) 0 (0.0) 0 (0.0)
227 None 36 (15.9) 36 (15.9) (0.0) 0 (0.0)
Nikouei, 2022 41 Aspirin - 0 (0.0) - -
41 None - 0 (0.0) - -
Rokito, 1996 75 None 0 (0.0) 0 (0.0) (0.0) -
35 Warfarin 0 (0.0) 0 (0.0) 2 (5.7) -
Shafiei, 2022 123 DOAC - - - 3 (2.4)
121 LMWH - - - 1 (0.8)
Voth, 1992 87 LMWH 4 (4.6) 4 (4.6) - -
92 Heparin 3 (3.3) 3 (3.3) - -
Yang, 2017 82 LMWH 0 (0.0) 0 (0.0) - -
82 None 4 (4.9) 4 (4.9) - -
Zhang, 2020 138 None 19 (13.8) 19 (13.8) 0 (0.0) 0 (0.0)
151 TXA 20 (13.2) 20 (13.2) 0 (0.0) 0 (0.0)
141 DOAC 3 (2.1) 3 (2.1) 0 (0.0) 0 (0.0)
169 TXA + DOAC 7 (4.1) 7 (4.1) 0 (0.0) 0 (0.0)

From all available studies, no group had a high probability of being the best or worst treatment in effectiveness for preventing VTE, demonstrating no indication of superiority (Fig. 8). Via the SUCRA plot, the warfarin group performed as the best treatment (61%) and the None (47%) and TXA (44%) groups performed as the worst treatments, but all having low probability (<75%). Other interventions also had a low probability of being the best or worst treatment. In the aspirin group (n = 50), 4.0% of patients experienced a VTE, compared to 2.4% in the DOAC group (n = 592), 3.4% in the heparin group (n = 117), 3.8% in the LMWH group (n = 693), 14.9% in the TXA group (n = 363), 3.9% in the TXA + DOAC group (n = 387), 0% in the warfarin group (n = 35), and 10.2% in the group that received no chemoprophylaxis (n = 596). However, when compared to the group that received no chemoprophylaxis (None), there was a statistically significant decrease in VTE occurrence in the aspirin, DOAC, LMWH, and TXA + DOAC groups (Fig. 8). There was minimal inconsistency for total VTE occurrence. There was “moderate” certainty of evidence based on study and grouping heterogeneity and risk of bias.

Incidence of venous thromboembolic events from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
Fig. 8 Incidence of venous thromboembolic events from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
3.5

3.5 Incidence of DVT by anticoagulant administered

From all available studies, no treatment arm had a high probability of being the best or worst treatment in preventing DVT, demonstrating no indication of superiority (Fig. 9). Via the SUCRA plot, the warfarin group performed as the best treatment (68%) and the TXA (34%) and None (33%) groups performed as the worst treatments, but these are reported as low probability (<75%). Other interventions had a low probability of being the best or worst treatment. In the aspirin group (n = 91), 2.2% of patients were diagnosed with a DVT, compared to 2.2% in the DOAC group (n = 592), 3.4% in the heparin group (n = 117), 3.0% in the LMWH group (n = 693), 14.9% in those that received TXA (n = 363), 3.9% in those that received TXA and a DOAC (n = 387), 0% in the warfarin group (n = 35), and 9.6% in patients that received no chemoprophylaxis (n = 637). However, when compared to the group that received no chemoprophylaxis (None), there was a statistically significant decrease in DVT occurrence in the aspirin, DOAC, LMWH, and TXA + DOAC groups (Fig. 9). There was minimal inconsistency for total VTE occurrence. There was “moderate” certainty of evidence based on study and grouping heterogeneity and risk of bias. There was minimal inconsistency for total DVT occurrence.

Incidence of deep vein thrombosis (DVT) from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
Fig. 9 Incidence of deep vein thrombosis (DVT) from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
3.6

3.6 Incidence epidural hematoma by anticoagulant administered

Across all studies, epidural hematoma was rare. From all available studies, no treatment arm had a high probability of being the best or worst treatment in respect to incidence of adverse epidural hematomas (Fig. 10). Via the SUCRA plot, the TXA + DOAC group performed the best (40%) and the DOAC group performed the worst (45%) at avoiding epidural hematomas, but these are reported as low probability (<75%). Other interventions had a low probability of being the best or worst treatment. In the DOAC group (n = 715), 0.6% of patients were diagnosed with an epidural hematoma, compared to 0.2% in the LMWH group (n = 555), and 0% in each the TXA group (n = 363), TXA and DOAC group (n = 387), and in the no chemoprophylaxis group (n = 365). Additionally, no interventions demonstrated a statistically significant difference in epidural hematoma incidence compared to the no chemoprophylaxis group (Fig. 10). There was “low” certainty of evidence based on study and grouping heterogeneity and risk of bias. There was minimal inconsistency for total epidural hematoma occurrence.

Incidence of epidural hematoma from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
Fig. 10 Incidence of epidural hematoma from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
3.7

3.7 Incidence bleeding complications by anticoagulant administered

From all available studies, no treatment arm had a high probability of being the best or worst treatment in preventing bleeding complications (Fig. 11). Via the SUCRA plot, the TXA + DOAC (38%) and TXA (34%) groups performed the best and the warfarin group performed the worst (61%) at avoiding bleeding complications, but these are reported as low probability (<75%). Other interventions had a low probability of being the best or worst treatment. In the aspirin group (n = 50), 4.0% of patients experienced an adverse bleeding event, compared to 4.1% in the DOAC group (n = 592), 28.0% in the heparin group (n = 25), 3.9% in the LMWH group, 5.7% in the warfarin group (n = 35), and 0% in both the TXA + DOAC group (n = 387) and the TXA group (n = 363). The heparin group experienced much greater rates of bleeding events; however, this can likely be attributed to the small sample size. Additionally, when compared those that received no chemoprophylaxis, only the warfarin group demonstrated a statistically significant increase in bleeding event incidence, however this too can be attributed to small sample size. There was “low” certainty of evidence based on study and grouping heterogeneity and risk of bias. There was minimal inconsistency for total adverse bleeding event incidence.

Incidence of bleeding complications from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
Fig. 11 Incidence of bleeding complications from all studies. In the Litmus Rank-O-Gram, higher surface under the ranking curve (SUCRA) values as well as higher cumulative probability curves represent better performance. The forest plot demonstrates risk ratio (RR) of each study group compared to the None group.
4

4 Discussion

This first-time NMA is the largest synthesis of evidence to date evaluating the effectiveness and safety of multiple chemoprophylactic agents against VTE and DVT following elective spine surgery. This study adds to the current literature by enhancing surgeon decision-making as to which agent, if any, is most appropriate for preventing VTE and DVT. Based on twelve RCTs, the results of this study cautiously suggest that there is not a high probability that one method of chemoprophylaxis is the best treatment for prevention of VTE in elective spine surgery. However, many of the medication classes failed to demonstrate a clinically or statistically significant result in preventing thromboembolic events when evaluated in head-to-head analyses. Importantly, no subgroup demonstrated high probability (>75%) of being the best or worst treatment for VTE or DVT prevention via the SUCRA plot. Safety was measured by adverse bleeding events, defined by author-reported significant/severe bleeding and incidence of epidural hematoma, and did not differ greatly between groups. Adverse bleeding events were similar among the remaining groups. Additionally, no treatment was deemed the best or worst with association to bleeding events via the SUCRA plot. Findings were similar when examining epidural hematoma, with incidence ranging from 0.0 to 0.6% among subgroups. Additionally, each group failed to produce a high probability of being the best or worst treatment for avoiding epidural hematoma occurrence. Notably, the mean operative time of the elective surgeries in this study was 169.0 (87.3) minutes, so the results of this study may not be generalizable to larger elective spine surgeries, warranting caution with data application.

Overall, our results suggest that one class of VTE chemoprophylactic agents may not be superior to other classes and may not be superior to mechanical or no prophylaxis. Thus, our results agree with those of Muralidharan et al. (2025) by adding to the literature that there is no overwhelming evidence that chemoprophylaxis outperforms no chemoprophylaxis in preventing thromboembolic events after elective spine surgery. Additionally, there was no significant safety benefit, in terms of adverse bleeding events, when comparing chemoprophylaxis and no chemoprophylaxis. The feared complication of providing anticoagulation after spine surgery is the development of an epidural hematoma due to the potential of rapidly progressive neurologic deficits.36 This study and Muralidharan et al. (2025) both demonstrated exceedingly low rates of epidural hematoma, ranging from 0 to 0.6% in all studies. This study adds to the current literature by further validating VTE risk and safety of different chemoprophylactic subgroups in elective spine surgery. However, caution should be exercised as these results may also be underpowered.

The North American Spine Society (NASS) recommends preventing thromboembolic events in all patients with mechanical prophylaxis in the form of compression stockings or PCDs, as well as early mobilization.6 NASS further maintains that chemoprophylaxis is to be considered in high-risk groups, but is not recommended for all patients due to the low incidence of VTE and risk of spinal epidural hematoma development following anticoagulation.37 The results of this study further corroborated the recommendations put forth in these guidelines in respect to elective spine surgery. Our findings of similar relative performance among subgroups are comparable to several recent systematic reviews and meta-analyses that evaluated VTE incidence in non-spine orthopedic surgery, which have shown non-inferiority of DOACs, LMWH, and aspirin.38–40 Similarly, this study reports similar findings to an elective spine surgery-specific meta-analysis demonstrating similar incidence of VTE, DVT, and adverse bleeding events when comparing chemoprophylaxis versus no chemoprophylaxis.9 It is important to note that VTE is a relatively rare postoperative complication in elective spine procedures, typically occurring in less than one percent of patients.2 The guidelines of the NASS, as well as the findings put forth by this study, are useful references for surgeon decision-making. However, physicians should consider VTE risk factors and comorbidities that may contraindicate certain medications, as well as type of surgical procedure when considering VTE prophylaxis. Advanced age, high body mass index (BMI), history of prior VTE, diabetes mellitus, and active malignancy are all well-documented risk factors for developing VTE after spine surgery.3,41 Additionally, trauma patients and those with spinal tumors have an increased risk of developing VTE.42 Lastly, it is well-documented that LMWH and certain DOACs should be avoided in patients with renal insufficiency.43 Applying the results of this study across populations should be done with caution, as the mean age of our study participants was 54 years old and mostly lumbar surgery, although some of this data was missing. This adds to the suggestion that surgeons must tailor their approach to VTE prophylaxis to unique patient factors in elective spine surgery.

4.1

4.1 Strengths and limitations

Strengths of this study include being the first NMA of RCTs to evaluate VTE and DVT incidence after elective spine surgery, corroborating the results of other systematic reviews and meta-analyses and enhancing surgeon's informed decision making. Another strength includes the comprehensive evaluation of both effectiveness and safety of a multitude of prophylactic regimens by examining VTE and DVT incidence as well as significant/severe bleeding and incidence of epidural hematoma, respectively. However, this study is not without limitations and warrants careful consideration. The inclusion and exclusion criteria greatly reduced the number of studies that qualified for inclusion, limiting generalizability and suggesting that analyses may be underpowered. Additionally, the constraints of this study pooled placebo, no intervention, and mechanical prophylaxis into one subgroup, increasing overall heterogeneity. In addition, inherently different surgeries and varying indications for surgery were grouped together and could not be stratified due lack of available data and difference in study design. For example, studies included in this NMA examined patients that underwent spinal decompression, discectomy, and internal fixation/fusion of different levels that could not be differentiated on a granular level. Similarly, timing of anticoagulant initiation varied among RCTs and could not be stratified due to lack of available data from authors and differences in study protocol. Furthermore, transitivity was generally acceptable, but a significant difference in operative time may have been present in one study. Additionally, several subgroups were only examined in one or two studies, yielding small sample sizes, particularly in the warfarin (n = 35), aspirin (n = 91), and heparin (n = 117) subgroups. As mentioned prior, comorbidities such as diabetes mellitus, chronic kidney disease, and BMI are independent risk factors for VTE and were not perfectly uniform in the included studies. Further research evaluating VTE prophylaxis in specific elective spine surgeries, i.e., decompression versus fusion, is warranted. There was also limited data available on the extent of the elective spine surgeries, such as the number of levels. However, this study likely focuses on relatively smaller elective spine surgeries rather than extensive multilevel fusions given the low mean operative time of 169.0 (87.3) minutes. Furthermore, large sample-size RCTs are warranted to directly compare effectiveness and safety of specific VTE chemoprophylactic agents following both short and long construct elective spine surgery.

5

5 Conclusion

This NMA of RCTs demonstrates that no chemoprophylactic agent has a high probability of being the best or worst treatment for preventing VTE, suggesting likely clinical equivalence. Similarly, no agent had a high probability of being the best or worst agent in terms of adverse bleeding events, including epidural hematoma. Surgeons should use this evidence cautiously and consider the use of VTE chemoprophylaxis following elective spine surgery on a case-by-case basis. These results appear consistent with prior literature examining the effectiveness and safety of VTE chemoprophylaxis. Further research would benefit from studying VTE chemoprophylaxis in specific elective spine surgery subtypes.

5.1

5.1 Statements and declarations

No funding was received to assist with the preparation of this manuscript. PGP has received research support from Cervical Spine Research Society (CSRS); has received other financial or material support from Allosource; is a paid presenter or speaker for Globus Medical; is a paid consultant for Medtronic, Royal Biologics, SpineWave, Terumo, and Zimmer; is on the editorial board for the Journal of Clinical Medicine (JCM) and Spine; and the senior chair editorial board for Journal of Neurosurgery: Spine. All remaining authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

Author contribution

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by NC, ANB, and SY. The first draft of the manuscript was written by NC and ANB and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript and agreed to be accountable for all aspects of the work.

Declarations

-No funding was received to assist with the preparation of this manuscript.-PGP reported the following relationships:oAllosource: other financial or material supportoCervical Spine Research Society: research supportoGlobus Medical: paid presenter or speakeroJCM: editorial boardoJNS Spine: senior chair editorial boardoMedtronic: paid consultantoRoyal Biologics: paid consultantoSpine: editorial or governing boardoSpineWave: paid consultantoTerumo: paid consultantoZimmer: paid consultant-The remainder of the authors have no relevant financial or non-financial interests.

Ethical statement

This study was reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Network Meta-Analyses (PRISMA-NMA). This NMA was also pre-registered on the International Prospective Register of Systematic Reviews (PROSPERO) on April 4th, 2025, prior to study initiation (CRD420251026334). No deviations from the initial protocol were made.

IRB approval was not sought because this study is a meta-analysis of previously published studies and did not involve new data collection or interaction with human participants.

Funding statement

No funding was received to assist with the preparation of this manuscript. PGP has received research support from Cervical Spine Research Society (CSRS); has received other financial or material support from Allosource; is a paid presenter or speaker for Globus Medical; is a paid consultant for Medtronic, Royal Biologics, SpineWave, Terumo, and Zimmer; is on the editorial board for the Journal of Clinical Medicine (JCM) and Spine; and the senior chair editorial board for Journal of Neurosurgery: Spine. All remaining authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

Credit author statement

NC: Methodology, Formal analysis, Investigation, Writing - Original Draft, Visualization.

ANB: Methodology, Formal analysis, Investigation, Writing - Original Draft, Visualization.

SY: Conceptualization, Software, Writing - Reviewing & Editing, Visualization.

EJC: Conceptualization, Writing - Reviewing & Editing, Supervision, Project Administration.

PGP: Conceptualization, Writing - Reviewing & Editing, Supervision, Project Administration.

DCG: Conceptualization, Writing - Reviewing & Editing, Supervision, Project Administration.

KTC: Conceptualization, Writing - Reviewing & Editing, Supervision, Project Administration.

JCH: Conceptualization, Writing - Reviewing & Editing, Supervision, Project Administration.

References

  1. , , , . Venous thromboembolism after spine surgery. Spine. 2014;39(11):911-918.
    [Google Scholar]
  2. , , , , , . Thromboembolic risk after lumbar spine surgery: a cohort study on 325 000 French patients. J Thromb Haemost JTH. 2018;16(8):1537-1545.
    [Google Scholar]
  3. , , , et al . Venous thromboembolic events after spinal fusion: which patients are at high risk? J Bone Joint Surg Am. 2014;96(11):936-942.
    [Google Scholar]
  4. , , , et al . Venous thromboembolism in adult elective spinal surgery: a tertiary centre review of 2181 patients. Bone Jt. 2017;99-B(9):1204-1209.
    [Google Scholar]
  5. , , , , , . Decreased incidence of venous thromboembolism after spine surgery with early multimodal prophylaxis: clinical article. J Neurosurg Spine. 2014;21(4):677-684.
    [Google Scholar]
  6. , , , et al . An evidence-based clinical guideline for the use of antithrombotic therapies in spine surgery. Spine J Off J North Am Spine Soc. 2009;9(12):1046-1051.
    [Google Scholar]
  7. , , , et al . The aging of the global population: the changing epidemiology of disease and spinal disorders. Neurosurgery. 2015;77(Suppl 4):S1-S5.
    [Google Scholar]
  8. , , , et al . Global, regional, and national burden of spine pain, 1990-2019: a systematic analysis of the global burden of disease study 2019. Arch Phys Med Rehabil. 2024;105(3):461-469.
    [Google Scholar]
  9. , , , et al . Venous thromboembolism chemoprophylaxis is not supported following elective spine surgery: a systematic review and meta-analysis of randomized controlled trials. J Spine Surg Hong Kong. 2025;11(2):242-255.
    [Google Scholar]
  10. , , , et al . Prevention of VTE in nonorthopedic surgical patients: Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e227S-e277S.
    [Google Scholar]
  11. , , , et al . Benefits and harms of direct oral anticoagulation and low molecular weight heparin for thromboprophylaxis in patients undergoing non-cardiac surgery: systematic review and network meta-analysis of randomised trials. Br Med J. 2022;376
    [Google Scholar]
  12. , , , . The PRISMA statement extension for systematic reviews incorporating network meta-analysis: PRISMA-NMA. Med Clínica Engl. 2016;147(6):262-266.
    [Google Scholar]
  13. , , , et al . Screening and prophylaxis for venous thromboembolism in pediatric surgery: a systematic review. J Pediatr Surg. 2024;59(10)
    [Google Scholar]
  14. , , , , . Rayyan—A web and mobile app for systematic reviews. Syst Rev. 2016;5(1):210.
    [Google Scholar]
  15. , , , et al . The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. Br Med J. 2011;343(oct18 2)
    [Google Scholar]
  16. , , , , . Bias in meta-analysis detected by a simple, graphical test. Br Med J. 1997;315(7109):629-634.
    [Google Scholar]
  17. , , , , , , . The schroth method for pediatric scoliosis: a systematic and critical analysis review. JBJS Rev. 2024;12(9)
    [Google Scholar]
  18. , , , et al . GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383-394.
    [Google Scholar]
  19. , , , , , . MetaInsight: an interactive web‐based tool for analyzing, interrogating, and visualizing network meta‐analyses using R‐shiny and netmeta. Res Synth Methods. 2019;10(4):569-581.
    [Google Scholar]
  20. , , , et al . Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: a systematic review. BMC Med Res Methodol. 2018;18(1):25.
    [Google Scholar]
  21. , , , , , . Imputing missing standard deviations in meta-analyses can provide accurate results. J Clin Epidemiol. 2006;59(1):7-10.
    [Google Scholar]
  22. , , , . A multifaceted graphical display, including treatment ranking, was developed to aid interpretation of network meta-analysis. J Clin Epidemiol. 2023;157:83-91.
    [Google Scholar]
  23. , , , et al . Effectiveness of various anterior noncontiguous cervical spine surgeries for treatment of noncontiguous cervical degenerative disk disease: a network meta-analysis. Asian Spine J. 2025;19(5):822-835.
    [Google Scholar]
  24. , , , , , , . Comparison of rivaroxaban and parnaparin for preventing venous thromboembolism after lumbar spine surgery. J Orthop Surg. 2015;10:78.
    [Google Scholar]
  25. , , , , . Prevention of thromboembolic complications with miniheparin-dihydroergotamine in patients undergoing lumbar disc operations. Eur Arch Psychiatry Neurol Sci. 1984;234(3):157-161.
    [Google Scholar]
  26. , , . Incidence of venous thromboembolic complications in instrumental spinal surgeries with preoperative chemoprophylaxis. J Korean Neurosurg Soc. 2015;57(2):114-118.
    [Google Scholar]
  27. , , , et al . Rivaroxaban versus enoxaparin as thromboprophylaxis in degenerative spine surgery: a randomized blinded non-inferiority study. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. 2025;34(5):1926-1933.
    [Google Scholar]
  28. , , , , , . A comparison between enoxaparin and aspirin in preventing deep vein thrombosis after spine surgery: a randomized clinical trial. Arch Bone Jt Surg. 2024;12(6):412-417.
    [Google Scholar]
  29. , , , et al . Combined use of tranexamic acid and rivaroxaban in posterior/transforaminal lumbar interbody fusion surgeries safely reduces blood loss and incidence of thrombosis: evidence from a prospective, randomized, double-blind, placebo-controlled study. Glob Spine J. 2023;13(5):1229-1237.
    [Google Scholar]
  30. , , , et al . Effect of aspirin in preventing deep vein thrombosis (DVT) after lumbar canal spinal stenosis surgeries: a double-blind parallel randomized clinical trial. Curr Orthop Pract. 2022;33(6):543-547.
    [Google Scholar]
  31. , , , . Deep vein thrombosis after major reconstructive spinal surgery. Spine. 1996;21(7):853-858.
    [Google Scholar]
  32. , , , et al . Comparison between rivaroxaban versus enoxaparin for venous thromboembolism prophylaxis following spine surgeries, a randomized clinical trial. J Clin Neurosci Off J Neurosurg Soc Australas. 2022;105:51-57.
    [Google Scholar]
  33. , , , , , , . Prevention of deep vein thrombosis in neurosurgical patients: a prospective double-blind comparison of two prophylactic regimen. Neurosurg Rev. 1992;15(4):289-294.
    [Google Scholar]
  34. , , , , , . Diagnosis and prevention of deep vein thrombosis after spine surgery in the elderly. Chin J Tissue Eng Res. 2017;21(23):3694-3699.
    [Google Scholar]
  35. , , , et al . Combined use of tranexamic acid and rivaroxaban in posterior lumbar interbody fusion safely reduces blood loss and transfusion rates without increasing the risk of thrombosis-a prospective, stratified, randomized, controlled trial. Int Orthop. 2020;44(10):2079-2087.
    [Google Scholar]
  36. , , , , . Postoperative spinal epidural hematoma: a systematic review. Spine. 2010;35(10):E413-E420.
    [Google Scholar]
  37. , , , et al . Risk factors and pharmacologic prophylaxis for venous thromboembolism in elective spine surgery. Spine J Off J North Am Spine Soc (6):970-978.
    [Google Scholar]
  38. , , , et al . Oral dabigatran versus enoxaparin for thromboprophylaxis after primary total hip arthroplasty (RE-NOVATE II*). A randomised, double-blind, non-inferiority trial. Thromb Haemost. 2011;105(4):721-729.
    [Google Scholar]
  39. , , et al . Aspirin or low-molecular-weight heparin for thromboprophylaxis after a fracture. N Engl J Med. 2023;388(3):203-213.
    [Google Scholar]
  40. , , , et al . Aspirin or rivaroxaban for VTE prophylaxis after hip or knee arthroplasty. N Engl J Med. 2018;378(8):699-707.
    [Google Scholar]
  41. , , , et al . Risk factors for and prediction of early thromboembolic disease following adult spinal deformity surgery: an analysis of >7,400 patients with spinal deformity. J Bone Joint Surg Am. 2025;107(11):1219-1226.
    [Google Scholar]
  42. , , , et al . Prevention of venous thromboembolism: the seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest. 2004;126(3 Suppl):338S-400S.
    [Google Scholar]
  43. , . Renal profiles of anticoagulants. J Clin Pharmacol. 2012;52(7):964-975.
    [Google Scholar]

Fulltext Views
34

PDF downloads
5
View/Download PDF
Download Citations
BibTeX
RIS
Show Sections

Suggested read for related articles: