Rivaroxaban versus Low Molecular Weight Heparin for Cancer-Associated Venous Thromboembolism: A Systematic Review and Meta-Analysis

Abstract

Introduction

Venous thromboembolism (VTE), which includes DVT and PE is a serious complication in cancer patients. This meta-analysis evaluates the efficacy and safety of rivaroxaban compared to LMWHs in treating cancer-associated VTE.

Methods

Following PRISMA guidelines, we searched PubMed, Cochrane Library, Google Scholar, and ClinicalTrials.gov from inception to May 2025 for studies comparing rivaroxaban with low-molecular-weight heparins (LMWHs) in adults with cancer-associated VTE. Primary outcomes were VTE recurrence and all-cause mortality. Risk ratios (RR) with 95% confidence intervals (CI) were pooled using a random-effects model. Risk of bias was assessed using RoB and NOS, with sensitivity and GRADE analyses performed.

Results

Twenty-five studies (n = 19,153, Rivaroxaban: 5186; LMWH: 13967) were included. Rivaroxaban significantly reduced the risk of VTE recurrence compared to LMWH (RR = 0.88; 95% CI: 0.78-0.99; P = 0.03). Rivaroxaban also significantly reduced all-cause mortality compared to LMWH (RR = 0.83; 95% CI: 0.69-0.99; P = 0.04). However, no statistically significant differences were noted in other outcomes. GRADE assessment showed evidence certainty ranged from high to very low.

Conclusion

Rivaroxaban demonstrates lower all-cause mortality compared with LMWH, indicating a potential advantage in managing cancer-associated VTE. However, further large-scale RCTs are warranted to confirm its safety profile.

1. Introduction

Venous thromboembolism refers to clot formation in different veins of body which includes deep vein thrombosis (DVT) and pulmonary embolism (PE), is a common and often life-threatening complication in cancer patients. ¹ This condition is a major cause of morbidity and the second leading cause of death in people with cancer. Moreover, the hypercoagulable state associated with cancer has been shown to contribute to tumor progression, independent of VTE events themselves. Hence also referred to as cancer-associated VTE (CAT). ²

The pathophysiology behind this event stems from the fact that cancer creates a hypercoagulable state through several mechanisms that increase the risk of blood clots. The coagulation cascade is being activated by tumor cells by releasing procoagulant substances, such as tissue factor (TF), which triggers a series of reactions that lead to clot formation. ³ Also, these active cancer cells secrete cytokines and other substances that can damage the blood vessel lining (endothelium), further promoting clot development. These events lead to formation of a thrombus which can either remain in place as a DVT or break off and travel to the lungs as a PE. Most observed symptoms of DVT include swelling, pain, tenderness, or redness in the affected limb, most often in the leg 4. Furthermore, a pulmonary embolism is more severe and acute event, with symptoms such as sudden shortness of breath, chest pain, cough, and rapid heartbeat. The severity of CAT varies, from asymptomatic or mild DVT to fatal PE.

VTE is generally more common in cancer patients then general population globally. The current epidemics of incidence of VTE is about 0.1% in the general population, it is estimated to be 0.5% in cancer patients, and some studies suggest that as many as 20% of all VTE events are cancer related. Cancer related factors that increase the risk of CAT include type of cancer, its stage, immobility and chemotherapy frequency. ⁵ One of the key risk factors contributing to the formation of DVT is obesity, obese people are more prone to formation of thrombus let alone DVT. ⁶

Conventional treatment of CAT has been anticoagulation therapy, specifically low molecular weight heparin (LMWH) like enoxaparin. This therapy is highly effective but require daily subcutaneous injections, which can be burdensome for patients, leading to reduced adherence and a potential for injection-site bruising. ⁷ Recently, a new direct oral anticoagulant (DOACs) like rivaroxaban have emerged as a more convenient, once-daily oral option. It is a direct Factor Xa inhibitor that prevents blood clot formation. Although convenient because of its oral administration, this anticoagulant, like all others, carries concerns regarding bleeding risk. ⁸

Cancer related DVT, is a serious complication that can lead to both acute and long-term health problems. The most frequent and critical threat is formation of pulmonary embolism when a part of DVT clot breaks off and travels to lungs via blood stream and lodges in a pulmonary artery blocking it. Ultimately leading to sudden shortness of breath, chest pain, and a rapid heartbeat. ⁹ In fact, after the cancer itself, PE is the second leading cause of death in cancer patients. Over half of cancer patients with proximal DVT may develop post-thrombotic syndrome (PTS). It is a chronic condition caused by damage to veins valves and linings. Resulting in impaired blood flow and stasis. Symptoms include persistent swelling, pain, heaviness, skin discoloration, and in severe cases, skin ulcers. This condition significantly reduces a patient’s quality of life and can require ongoing medical management. ¹⁰

The aim of this study is to compare the efficacy and safety of rivaroxaban versus conventional LMWH therapy in patients with cancer-associated VTE. Our findings will help determine if rivaroxaban can provide a safer, more effective, and more patient-friendly alternative to conventional LMWH therapy for this at-risk population.

2. Methods

This meta-analysis was reported following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. ¹¹ The protocol of this meta-analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO) with the reference number CRD420251169954.

2.1. Data Sources and Search Strategy

A thorough search was carried out using PubMed/MEDLINE, Embase, Scopus, Web of Science, and Google Scholar databases. The reference lists of the retrieved publications, as well as previous systematic reviews and meta-analyses, were examined for relevant articles. The search method placed no constraints on publishing status or language. The search words included relevant MeSH terms and associated keywords, such as (“venous thromboembolism” OR “VTE” OR “pulmonary embolism” OR “PE” OR “deep vein thrombosis” OR “DVT” OR “cancer-associated thrombosis” OR “CAT”) AND (“rivaroxaban” OR “apixaban” OR “DOAC” OR “direct oral anticoagulant”) AND (“dalteparin” OR “enoxaparin” OR “LMWH” OR “low molecular weight heparin”). Although broader direct oral anticoagulant (DOAC) terms were included in the search strategy to maximize sensitivity and ensure comprehensive study identification, only studies directly comparing rivaroxaban with low-molecular-weight heparin (LMWH) were eligible for inclusion in the final quantitative analysis. The detailed search strategy will be provided in Supplementary Table 1.

2.2. Study Selection and Eligibility Criteria

All articles retrieved through the systematic search were imported into the EndNote reference library, version X8.1 (Clarivate Analytics), where duplicates were thereafter omitted. Two authors independently examined and chose studies, and any differences were handled by a third author. Selected studies were retrieved for full-text review to ensure relevance. The inclusion criteria were defined using the Population, Intervention, Control, and Outcomes (PICO) methodology for systematic reviews, where:

• ● P (Population): Patients with active cancer and venous thromboembolism (VTE) or cancer-associated thrombosis (CAT).

• ● I (Intervention): Rivaroxaban (or other direct oral anticoagulants [DOACs], such as apixaban).

• ● C (Comparator): Low-molecular-weight heparin (LMWH), typically dalteparin or enoxaparin, or other DOACs.

• ● O (Outcomes): Key efficacy outcomes included recurrent VTE (symptomatic, incidental DVT, PE, or worsening of vascular obstruction). Key safety outcomes included major bleeding, clinically relevant non-major bleeding (CRNMB), and all-cause mortality.

This meta-analysis included randomized controlled trials and prospective or retrospective cohort studies that compared rivaroxaban or other DOACs with LMWH or other anticoagulants for the treatment or prevention of VTE/CAT in cancer patients. Studies explicitly focusing on high bleeding risk patients on DOACs or specific cancer types (e.g., luminal gastrointestinal or genitourinary cancers in some studies) were noted as relevant for subgroup analysis or exclusion criteria if specified by the individual study.

2.3. Data Extraction

On a pre-piloted Microsoft Excel sheet, two authors independently evaluated the data and supplemental resources; disagreements were settled by consulting a third author. From the selected studies, the following information was extracted: study labels, year of publication, study design, country, number of participants in each arm, baseline patient characteristics (e.g., age, sex, cancer type, VTE type, presence of metastases), duration of follow-up, and results about the effectiveness and safety profile. Effectiveness endpoints included recurrent VTE (overall, symptomatic, incidental), while safety endpoints included major bleeding, clinically relevant non-major bleeding, and all-cause mortality.

2.4. Quality Assessment

Two authors independently assessed the quality and risk of bias of the included studies. For Randomized Control Trials (RCTs), the Cochrane Risk of Bias tool 2.0 (RoB 2.0) was used. ¹² This tool assesses the risk of bias associated with the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Judgments were categorized as ‘Low risk’, ‘High risk’, or ‘Some concerns’. For observational studies, Newcastle and Ottawa Scale was utilized, focusing on aspects such as patient selection, comparability of groups (e.g., confounding control methods like propensity score matching), ascertainment of exposures and outcomes, and completeness of follow-up. ¹³

2.5. Statistical Analysis

Two authors conducted the integration of study findings using Review Manager (RevMan version 5.4; Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). For dichotomous outcomes (e.g., VTE recurrence, major bleeding, mortality), we calculated pooled Relative Risks (RR) with their 95% confidence intervals (CI). For continuous outcomes, Standardized Mean Differences (SMDs) with 95% CI were computed. We combined the findings from all included research using the Mantel-Haenszel random-effects model, and we displayed them in forest plots. A p-value of less than 0.05 was considered statistically significant. Utilizing the Higgins I² index, statistical heterogeneity was evaluated. High heterogeneity is indicated by I² > 50%, moderate by I² 25–50%, and low by <25%. Funnel plots were used to assess publication bias for outcomes reported by at least 10 studies.

2.6. Certainty of Evidence

We used a GRADE (Grading of Recommendations Assessment, Development, and Evaluation) tool to assess the overall quality and strength of evidence for each primary and secondary outcome. ¹⁴

3. Results

3.1. Literature Search Results

The initial literature search through three data bases and one register, yielded a total of 306 studies. After exclusions, 25 studies remained for analysis.^15-39 The PRISMA flowchart (figure-1) summarizes the process of literature search and screening.

Figure 1.

PRISMA 2020 flow diagram for study selection

3.2. Study Characteristics

This review includes 4 randomized controlled trials and 21 observational studies. The 25 included studies reported data on a total of 19,153 patients with cancer-associated venous thromboembolism (VTE). Out of these, 5186 patients received the intervention, rivaroxaban and 13,967 patients were assigned to the control group, LMWH. Follow-up duration ranged from 2 weeks to 24 months and the mean age of participants ranged from 60 to 78 years with varying standard deviations across the studies. Other baseline characteristics of all studies are summarized in (table1).

Table 1.

Study Characteristics and Baseline Patient Demographics

Study	Type of study	Number of patients	Inclusion criteria	Exclusion criteria	Follow up	Age years		BMI Kg/m2			Cancer sites in patients given rivaroxaban
						Rivaroxaban	Control	Rivaroxaban	Control	Active cancer patients (n)	Breast	Lung	Gastrointestinal	Ovarian	Hematological	Pancreatic
Prins et al., 2014 15	Sub-group analysis of two Randomised controlled Trials.	8282	Patients with cancer from the EINSTEIN-DVT and EINSTEIN-PE trials (two randomised, open-label, phase 3 studies) were included. Patients were potentially eligible for the EINSTEIN-DVT and EINSTEIN-PE studies if they had symptomatic deep-vein thrombosis or pulmonary embolism.	The main exclusion criteria were a therapeutic dose of low-molecular-weight heparin, fondaparinux,or unfractionated heparin formore than 48 h before randomisation; more than a single dose of a vitamin K antagonist; treatment of the present episode with thrombectomy, a caval filter, or fi brinolytic therapy; any contraindication listed in the local labelling of enoxaparin, warfarin, or acenocoumarol; a creatinine clearance less than 30 mL/min; pregnancy or breastfeeding; active bleeding or a disorder at high risk for bleeding; or a life expectancy of less than 3 months	12 months	65-75	65-75	27·4 (5·4)	26·7 (4·6)	655	NA	NA	NA	NA	NA	NA
Ageno et al., 2017 16	Retrospective chort study	587	All adult inpatients and outpatients referred to the thrombosis unit with clinically suspected venous thromboembolism (VTE), confirmed diagnosis of deep vein thrombosis (DVT) or pulmonary embolism (PE) through objective testing.	Patients with superficial vein thrombosis or thrombophlebitis, unusual site thromboses such as splanchnic vein thrombosis, retinal vein thrombosis, and cerebral vein thrombosis, patients without an objective diagnosis of VTE, arterial thrombosis such as stroke, acute myocardial infarction, or peripheral arterial disease, recurrent abortion, pregnancy complications, and VTE prophylaxis.	rivaroxaban alone or after heparin or fondaparinux for ≤48 h. early switchers (rivaroxaban after heparin or fondaparinux for >48 h to 14 days and/or a VKA for 1–14 days); standard anticoagulation (heparin or fondaparinux and a VKA); LMWH alone; and miscellaneous (other heparins, fondaparinux alone, VKA alone).	66.2 (10.6)	66.2 (10.6)	27.9 (5.3)	27.9 (5.3)	587	32 (21.9)	5 (3.4)	20 (13.7)	NA	12 (8.2)	NA
Nicklaus et al., 2017 17	Retrospective Comparitive Study	90	Eligible patients were prescribed rivaroxaban or enoxaparin between 1 January 2012 to 31 August 2015 and followed up in clinic within 90 days initiation of therapy.	Patients were excluded from the study if they were prescribed a concomitant anticoagulant other than enoxaparin or rivaroxaban (i.e., heparin, warfarin, dabigatran, apixaban, or edoxaban), were less than 18 years of age, diagnosed with a thrombophillic condition, lacked adequate documentation of initial VTE event, or received rivaroxaban or enoxaparin at an inappropriate dose for VTE management	3 months	57.9	59.4	NA	NA	90	NA	NA	NA	NA	NA	NA
Alzghari et al., 2017 18	Retrospective Comparitive Study	127	Patients 18 years or older with active cancer diagnosis and a VTE with an expected three-month follow-up period were included. Patients had to be receiving enoxaparin, oral Xa inhibitor, or warfarin for at least 75% of total time on anticoagulants	Patients were excluded if they entered hospice, were pregnant, received systemic fibrinolytic therapy, had known adherence problems, experienced an underlying clotting or bleeding disorder, received a vena cava filter, received a mechanical heart valve, or received primary prophylactic anticoagulants.	3 months	62 ± 13.8	62.4 ± 13.9	NA	NA	127	9 (17)	8 (17)	NA	7 (15)	NA	NA
Chaudhury et al., 2017 19	Retrospective chart review	286	Patients were included if they were at least 18 years of age, had a diagnosis of cancer and concurrent diagnosis of DVT or PE, and were prescribed treatment with either rivaroxaban or dalteparin during the study period	Patients were excluded if the length of anticoagulation therapy was 30 days, if therapy with dalteparin or rivaroxaban was initiated more than 6 months after DVT or PE diagnosis, or if they did not receive therapeutic doses of therapy	upto 6 months	62.20 ± 12.34	59.31 ± 12.02	29.04 ± 5.78	28.59 ± 6.15	286	NA	NA	NA	NA	19 (19.6%)	NA
Streiff et al., 2018 20	A retrospective longitudinal cohort design	2428	Newly diagnosed patients with cancer (defined as at least one inpatient stay or two outpatient visits with a diagnosis of cancer with a prior washout period of 6 months) and with a first VTE (index VTE) after 2013 were identified. VTE can be an early sign of cancer, so the index VTE had to occur within a 30‐day window before the first cancer diagnosis or at any time after the first cancer diagnosis.	Patients with a prior VTE diagnosis or with an anticoagulant (i.e., warfarin, rivaroxaban, LMWH, unfractionated heparin, fondaparinux, apixaban, dabigatran, and edoxaban) dispensing before their index VTE were excluded from the study.	6 to 12 months	72.7 ± 8.8	72.6 ± 10.8	NA	NA	2428	NA	NA	NA	NA	NA	NA
Young et al., 2018 21	Randomized Trial	406	Adults aged 18 years and over with active cancer (diagnosed or treated within 6 months), objectively confirmed first episode of VTE (DVT or PE, symptomatic or incidental), ECOG less than and equal to 2, weight > 40 kg, and adequate organ function.	Prior VTE or full-dose anticoagulation before randomization, high bleeding risk, active bleeding, uncontrolled hypertension, or liver disease, concurrent use of strong CYP3A4/P-gp modulators or contraindicated medications.	6 months from randomization to primary outcome assessment	67	67	26.7	26.6	406	NA	NA	NA	NA	NA	NA
Signorelli et al., 2019 22	Retrospective pilot analysis/Comparative study	49	patients with gynecologic malignancies who received either rivaroxaban, warfarin, or LMWH for treatment of VTE (DVT or PE, first or recurrent).	Other NOACs were not included as they have not been actively prescribed to the gynecologic oncology patients at our medical center	6 months	60.4 ± 12.7	60.4 ± 13.1	NA	NA	49	NA	NA	NA	NA	NA	NA
Wysokinski et al., 2019 23	Retrospective Observational Study	750	Consecutive patients were evaluated with the diagnosis of acute Ca-VTE, they received either rivaroxaban, apixaban, or LMWH within the first 14 days after diagnosis, and completed at least 3 months of anticoagulant therapy, or had an outcome event in that period, and had an active cancer	Patients that did not have active cancer and were not talking any of the three specified anticoagulants.	3 months	62 ± 13	62 ± 12	NA	NA	750	12 (7.4)	17 (10.4)	34 (20.9)	7 (4.3)	10 (6.1)	21 (12.9)
Yang et al., 2019 24	Retrospective Observational Study	72	​	​	6 months	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA
Shoutian et al., 2019 25	Comparative Observational Study	394	Adults aged ≥ 18 years old, diagnosed with one of the following cancers: gastric, lung, esophageal, breast, colorectal, or ovarian, scheduled for treatment via PICC insertion	Overt predisposition to bleeding: e.g. platelets < 60×109/L; prothrombin time > 16s; aPTT > 50s; thrombin time > 22s; fibrinogen < 2.0 g/L, history of thrombosis with previous thrombolytic therapy, obvious dysfunction of heart, liver, or kidney, failure of PICC indwelling.	Catheter indwelling time ranged from 1 to 160 days	57.2 ± 9.7	56.7 ± 7.7	22.5 ± 0.2	22.6 ± 0.2	394	NA	NA	NA	NA	NA	NA
Lee et al., 2019 26	Retrospective Cohort Study	204	The study inclusion criteria were diagnosis of primary lung cancer, diagnosis of PE and/or DVT, and treatment of the VTE with rivaroxaban or dalteparin between January 1, 2012 and December 31, 2016.	exclusion criteria included rivaroxaban or dalteparin use for 7 days or less, use of other anticoagulants before rivaroxaban or dalteparin for >7 days, follow-up loss after the 1st visit, and inadequate dosing.	all causes of mortality were analyzed until March 31	66.79±11.32	64.25±8.66	NA	NA	204	NA	Adenocarcinoma: 85 (64.9) Squamous cell carcinoma: 21 (16) Small cell carcinoma: 18 (13.7) Large cell carcinoma: 1 (0.8)	NA	NA	NA	NA
Lee et al., 2020 27	Retrospective Cohort Study	162	diagnosis of ovarian, cervical, uterine, vaginal, vulvar, endometrial, or fallopian tube cancer, presence of VTE, as diagnosed via computed tomography or lower extremity Doppler ultrasonography by certificated radiologists, and treatment of VTE with a standard dose of rivaroxaban or dalteparin	Anticoagulant therapy initiated at another Rivaroxaban for VTE in gynecologic cancer, rivaroxaban or dalteparin use for ≤7 days without bleeding events, patients lost to follow-up after the first visit, and venous thromboembolic lesions removed by surgery	all causes of mortality were analyzed until March 31	56.90±12.32	56.27±10.34	NA	NA	162	NA	NA	NA	Ovarian (non-clear cell type): 35 (34.3) Ovarian (clear cell type): 8 (7.8)	NA	NA
Faqah et al., 2020 28	Retrospective Cohort Study	150	Adults diagnosed with cancer-associated VTE (DVT or PE), treated with either enoxaparin or rivaroxaban, radiographically confirmed VTE, symptomatic presentation of VTE	Patients with contraindications to anticoagulation therapy, pregnant or breastfeeding women, patients with active bleeding or high bleeding risk, severe renal impairment (e.g., creatinine clearance <30 mL/min), known hypersensitivity to enoxaparin or rivaroxaban	6 months	51.84 ± 13.49	48.67 ± 14.45	25.20 ± 5.43	23.63 ± 4.64	135	NA	NA	NA	NA	NA	NA
Costa et al., 2020 29	Retrospective Cohort Study	1058	Adults diagnosed with active (primary or metastatic) breast, lung, ovarian, or pancreatic cancer, who were admitted to the hospital or attended the emergency department and suffered from VTE, who had >12 months of continuous medical and prescription benefits before the CAT, and who received rivaroxaban or an LMWH as their first outpatient anticoagulant for the index event.	Patients without active cancers and not receiving any outpatient anticoagulant were excluded	12 months	NA	NA	NA	NA	1058	36.9	43.5	NA	6.2	NA	10.2
Chen et al., 2021 30	Retrospective Cohort Study	99	patients with active cancer who had a diagnosis of acute VTE (venous thromboembolism), they must have received LMWH (dalteparin) as initial therapy and then been switched to rivaroxaban successfully	Patients who died during treatment with LMWHs or had contraindications for rivaroxaban and patients who had received rivaroxaban treatment for less than one month were excluded	6 months	67 (34-84)	NA	21.6+1.4	NA	99	9 (9.1%)	13 (13.1%)	24 (24.2%)	NA	NA	NA
Houghton et al., 2021 31	Prospective cohort	1392	Consecutive patients with acute Ca-VTE treated at the Thrombophilia Clinic, Gonda Vascular Center, Mayo Clinic Rochester between March 1, 2013, and April 20, 2020, were included in this study	patients without GI or non-GI cancer (i.e. patients without cancer) are excluded by design (only Ca-VTE), the study does not in the abstract list detailed laboratory, renal, bleeding risk, or comorbidity exclusions, it is implicit that patients must have had sufficient follow-up and standard care at Mayo Clinic to be observed prospectively (i.e., exclusion of those lost to follow-up or incomplete data)	3 to 6 months	62.0 (12.4%)	62.4 (11.9%)	NA	NA	1392	NA	NA	NA	NA	NA	NA
Planquette et al., 2022 32	Randomized Trial	158	Consecutive adult patients with solid cancer, high-grade lymphoma orthalidomide, or lenalidomide-treated myeloma with a newly diagnosed symptomatic or incidental proximal lower-limb DVT, symptomatic orincidental iliac or inferior vena cava thrombosis or PE, or both and a high risk of recurrent VTE despite anticoagulation as estimated by a modified Ottawa score of >1were eligible for enrolment in the study.	The main exclusion criteria were adjuvant hormonal treatment with nomeasurable residual disease, isolated subsegmental PE withoutassociated proximal DVT, isolated distal DVT of the legs, isolatedupper-extremity DVT or superior vena cava thrombosis, isolatedvisceral thrombosis, platelet count of <50 G/L, active bleeding,hepatic disease associated with coagulopathy, and clinically relevantbleeding risk, including patients with cirrhosis with Child-Pughscores B and C, vena cava filter at inclusion, creatinine clearanceof <30 mL/min according to Cockcroft-Gault formula,anticoagulant treatment at curative dosage for more than 3 or 4daysbefore inclusion, pregnancy or lack of effective contraceptivetreatment for women of childbearing age, treatment with strongCYP3A4 or P-glycoprotein inhibitors, treatment with strongCYP3A4 inducers, estimated life expectancy <3 months, andEastern Cooperative Oncology Group score of 3 or	3 months	68.6 (62.9-77.8)	70.7 (62.7–78.7)	NA	NA	158	6 (8.1)	15 (20.3)	NA	6 (8.1)	Myeloma or Lymphoma: 7 (9.5)	6 (8.1)
Kim et al., 2022 33	Randomized Controlled phase 2 Trial	90	patients aged ≥19 and <80 years old, with histologically confirmed, advanced upper GI tract, hepatobiliary, and pancreatic cancer, and newly diagnosed (within 2 weeks before randomization) symptomatic or incidental lower extremity or upper extremity (jugular, innominate, subclavian, axillary, brachial) deep vein thrombosis (DVT) or pulmonary thromboembolism (PTE).	The major exclusion criteria included taking anticoagulant therapy for more than three days prior to randomization, any dose of thrombolysis therapy, continuous antiplatelet agent use, a history of VTE, hemodynamic unstable PTE, current active bleeding or uncontrolled bleeding history within 4 weeks before randomization, liver cirrhosis (Child–Pugh score ≥ 7), uncontrolled hypertension, and brain metastasis.	6 months	64 (39 to 77)	63 (42 to 78)	22.5 ± 3.3	22.4 ± 3.1	90	NA	NA	NA	NA	NA	6 (13.6)
Song et al., 2023 34	Retrospective Cohort Study	466	Patients were included if they were aged 18 years or older and had a diagnosis of active lung cancer and acute symptomatic or incidental PE with or without DVT after objectively confirmation.	Patients with high-risk PE and hemodynamic instability were excluded, as well as patients with thrombectomy, vena cava filter placement or thrombolysis for acute episode, with a therapeutic dose of any anticoagulants during the 12-month pre-index period, and with anticoagulation therapy less than seven days. Patients with other active cancer or previous chronic thromboembolic disease were excluded	12 months	65 (61,71)	65 (59, 70)	NA	NA	466	NA	Adenocarcinoma: 170 (78.7) Squamous cell carcinoma:28 (13.0) Small cell lung cancer: 6 (2.8)	NA	NA	NA	NA
Coleman et al., 2023 35	Retrospective Cohort Study	3,708	Adult patients with active primary or metastatic cancer excluding esophageal, gastric, unresected colorectal, bladder, and noncerebral central nervous system cancers and leukemia who were admitted to the hospital, emergency department, or observation unit for acute deep vein thrombosis. Patients had to be active in the EHR data set for at least 12 months before the index event and had to have at least 1 provider visit in the 12 months before the acute VTE event (baseline period).	patients from this analysis were excluded if they had atrial fibrillation or valvular heart disease, were pregnant, or were using anticoagulation during the baseline period.	6 to 12 months	Age <40, y: 3.8 Age 41-60, y: 31.2 Age 61-74, y: 40.4 Age >75, y: 24.6	Age <40, y: 3.8 Age 41-60, y: 30.3 Age 61-74, y: 40.7 Age >75, y: 25.2	BMI <29, kg/m2: 55.7 BMI 30-34, kg/m2: 23.4 BMI 35-39, kg/m2: 11.3 BMI >40, kg/m2: 9.5	BMI <29, kg/m2: 62.0 BMI 30-34, kg/m2: 19.3 BMI 35-39, kg/m2: 9.8 BMI >40, kg/m2: 8.8	3,708	25.3	18.3	Upper gastrointestinal: 2.6	3.8	Myeloma: 3.3 Lymphoma 9.4	3.6
Khan RAA et al., 2023 36	Retrospective study	250	patients with active cancer and associated acute VTE wre included	patients with chronic illness were excluded	12 months	NA	NA	NA	NA	NA	250	NA	NA	NA	NA	NA
Linder et al., 2024 37	Retrospective Cohort Study	5464	Patients in Sweden with a cancer diagnosis between 2013 and 2019, followed by a venous thromboembolism (VTE) diagnosis within 6 months (i.e. index CAT), Age ≥ 18 years at the time of the index VTE, living in Sweden for at least 183 days before the index VTE (so there is baseline exposure data), first dispense of rivaroxaban or LMWH within 28 days after the index VTE.	Patients with diagnoses or treatments in the 183 days before the index VTE that indicate other anticoagulation indications: atrial fibrillation, hip/knee replacement, acute coronary syndrome, or prior use of DOAC, vitamin K antagonists or LMWH, cancer types considered high bleeding risk according to ISTH guidance were excluded, e.g. lip/upper gastrointestinal cancers, malignant immunoproliferative diseases, leukemia, non-melanoma skin cancer	2013 to 2019	66	66	NA	NA	5464	30 (11%)	NA	NA	NA	NA	NA
Dong et al., 2024 38	Retrospective Cohort Study	296	Previously diagnosed with malignant tumors, diagnosis of VTE confirmed by CT and/or ultrasonography, anticoagulation therapy with rivaroxaban or LMWH, Age≥ 18 years old, Body weight≥ 40 kg, The patient’s renal function test value was within the normal range when VTE was diagnosed, There were no abnormalities in the patient’s blood routine and liver function tests that affected the antitumor treatment plan at the time of diagnosis of VTE.	Within 3 months before the date of enrollment, there is a clear record of the use of anticoagulants, or aspirin (>75 mg daily), clopidogrel (>daily). 50 mg), Patients have medical records of confirmed VTE before the date of enrollment, Have a record of diagnosis related to liver disease within 3 months before the date of enrollment, Diagnosis of bacterial endocarditis within 3 months prior to the date of enrollment, Diagnosis of major bleeding or clinically relevantnonmajor bleeding (CRNMB) within 3 months before the date of enrollment, The patient has been pregnant within 3 months before the date of enrollment, Patients have medication records of cytochrome P4503A4 inhibitors, P-glycoprotein inhibitors, or inducers from the date of enrollment to 24 months thereafter	6 months	70(62,74)	68(60,73)	36(36)	31(31)	296	NA	NA	NA	NA	NA	NA
Cohen et al., 2025 39	Retrospective Cohort Study	2259	adults ≥ 18 years with active cancer and a first cancer-associated VTE (CT), received therapeutic rivaroxaban or LMWH within 30 days after the CT diagnosis (or discharge), eligible patients had ≥ 1 year of medical history in CPRD/HES data before the CT to assess baseline covariates.	Prior history of VTE, prior therapeutic anticoagulation, or other indications for long-term anticoagulation (e.g. atrial fibrillation, mechanical heart valves), patients with contraindications to rivaroxaban (e.g. conditions with high bleeding risk) or stratified to exclude certain cancer types not recommended for DOAC use according to ISTH guidance (e.g. unresected colorectal, GI bleed–prone cancers), patients with insufficient prior data (less than 1-year prior history), palliative care initiation before CT, pregnancy, severe renal disease, or switching anticoagulant before at-risk period start.	12 months	72.4 (12.1)	66.9 (11.7)	27.8 (5.4)	28.1 (5.7)	2259	104 (33.1)	29 (9.2)	16 (5.1)	6 (1.9)	NA	NA

3.3. Risk of Bias Assessment

Risk of bias for included randomized control trials was assessed using RoB 2.0. The overall risk of bias for four trials was categorized as ‘some concerns’. One trial was classified as low risk, while the other three were assessed to have some concerns. The most common domain that indicated issues was ‘bias due to deviation from intended interventions’ (D2). All other domains showed a low risk of bias (Supplementary Figure S1). In addition to that, to assess the quality of observational studies, Newcastle-Ottawa Scale (NOS) was used. 9 studies scored 8 stars and 8 studies scored 7 stars suggesting high methodological quality and a low risk of bias. The remaining 4 studies scored 6 stars reflecting moderate quality and moderate risk of bias. The most common domain for the risk was the ‘comparability domain’ which assessed the control for confounding factors. The detail of NOS scores is provided in the Supplementary Table 2. Overall, the included studies demonstrated satisfactory methodological quality and support the validity of our findings.

3.4. Certainty of Evidence

The certainty of evidence for comparing Rivaroxaban with low molecular weight heparin (LMWH) in patients with venous thromboembolism (VTE) was assessed using the GRADE framework. Overall, the quality of evidence ranged from high to very low. High-certainty evidence supported the outcome of VTE recurrence, where Rivaroxaban significantly reduced the risk compared with LMWH. In contrast, evidence for DVT recurrence and PE recurrence was rated as low certainty, while evidence for major bleeding and all-cause mortality was of very low certainty. The details of grade assessment are presented in Supplementary Table 3.

3.5. Results of Meta-Analysis

3.5.1. Recurrence of Venous Thromboembolism

Nineteen studies reported recurrence of venous thromboembolism (VTE) as an outcome with 3489 patients receiving Rivaroxaban and 10608 receiving LMWH. The pooled analysis shows that Rivaroxaban was associated with significantly lower risk of VTE recurrence compared to low molecular weight heparin (LMWH) [RR 0.88, 95% CI 0.78-0.99; P=0.03; I² = 0%] as demonstrated in Figure 2.

Figure 2.

Forest plot showing recurrence of VTE

3.5.2. Recurrence of Deep Vein Thrombosis

Seventeen studies reported recurrence of deep vein thrombosis (DVT) as an outcome, including 3119 patients in the Rivaroxaban group and 5823 in the LMWH group. The forest plot indicates no statistically significant difference in DVT recurrence between Rivaroxaban and LMWH [RR=0.96, 95% CI= 0.73 to 1.26; P=0.75, I² = 29%] as demonstrated in Figure-3.

Figure 3.

(A) Forest plot showing Recurrence of Deep Vein Thrombosis. (B) Sensitivity analysis forest plot showing Recurrence of Deep Vein Thrombosis

3.5.3. Recurrence of Pulmonary Embolism

Seventeen studies reported recurrence of pulmonary embolism (PE) as an outcome. 3183 patients were in the intervention group while 5882 served as controls. According to the results presented in the forest plot (figure-4), no statistically significant difference was found in recurrence of pulmonary embolism (PE) among patients treated with rivaroxaban and those who were treated with LMWH [RR= 0.93, 95% CI= 0.77 to 1.13; P=0.48, I² = 11%].

Figure 4.

(A) Forest plot showing Recurrence of Pulmonary Embolism. (B) Sensitivity analysis forest plot showing Recurrence of Pulmonary Embolism

3.5.4. Major Bleeding

Twenty studies that included 4074 patients receiving Rivaroxaban and 12010 receiving LMWH, reported major bleeding as an outcome. According to our analysis (Figure-5), rivaroxaban and low molecular weight heparin (LMWH) were found to be comparable in terms of major bleeding and showed no statistically significant difference [RR =1.01, 95% CI= 0.79 to 1.29; P=0.94, I²= 27%].

Figure 5.

(A) Forest plot showing Major Bleeding. (B) Sensitivity analysis forest plot showing Major Bleeding

3.5.5. All - Cause Mortality

Twelve studies reported all-cause mortality as an outcome, including 1729 patients in the Rivaroxaban arm and 3156 in the LMWH arm. Figure-6 demonstrates the primary analysis, which shows that Rivaroxaban is associated with a statistically significant reduction in all-cause mortality compared to low molecular weight heparin (LMWH) [RR 0.83, 95% CI 0.69-0.99; P=0.04, I²=53%]. However, following a sensitivity analysis, that excluded Song et al. 2023 and Cohen et al. 2025, the association of Rivaroxaban to lower incidence of all–cause mortality, was no longer significant, with a marked drop in heterogeneity from 53% to 5%[RR 0.95, 95% CI 0.84 to 1.07; P=0.40, I²=5%] (Figure S3).

Figure 6.

(A) Forest plot showing All-Cause Mortality. (B) Sensitivity analysis forest plot showing All-Cause Mortality

3.6. Publication Bias

We assessed publication bias across all outcomes using funnel plots. For VTE recurrence, the plot showed mild asymmetry, with fewer small studies on the right side (RR > 1) compared to the left, suggesting possible underrepresentation of studies reporting increased risk. The funnel plot for DVT recurrence also appeared asymmetric, with a clustering of small studies on the left side (RR < 1) showing reduced recurrence, raising concern for selective reporting of protective findings. For PE recurrence, asymmetry was noted, with smaller studies clustering on the right side (RR > 1), indicating potential publication bias towards increased risk. The plot for major bleeding demonstrated slight asymmetry, with more studies on the right side (RR > 1) showing increased bleeding risk, which may indicate minimal underrepresentation of neutral or protective findings. Finally, the funnel plot for all-cause mortality showed clear asymmetry, with fewer small studies on the left side (RR < 1) reporting reduced mortality, consistent with possible publication bias. Funnel plots for each outcome are provided in the Supplementary Figures S4-S8.

4. Discussion

Through this systematic review and meta-analysis, we aimed to compare rivaroxaban with low molecular weight heparin (LMWH) for the treatment of venous thromboembolism (VTE), focusing on recurrence, bleeding risk, and mortality outcomes. While previous meta-analyses have primarily concentrated on rivaroxaban in cancer-associated thrombosis or pooled it together with other direct oral anticoagulants,^40-42 to our knowledge, no prior meta-analysis has comprehensively evaluated rivaroxaban against LMWH across all primary VTE outcomes, including recurrence, bleeding, and mortality.

Venous thromboembolism (VTE) remains a major global health problem and is a leading cause of morbidity and mortality worldwide.^43,44 In addition to the immediate risks of acute thrombosis, patients with VTE also face long-term complications, for example, recurrent events, post-thrombotic syndrome, and chronic thromboembolic pulmonary hypertension, which impair quality of life and increase medical expenses. ⁴⁵ They are usually prescribed with initial anticoagulation therapy; however, the risk of recurrence is high. Therefore, there exists a need for the development of effective and sustainable treatment strategies. Low molecular weight heparin (LMWH) has long been regarded as the foundation of initial anticoagulation therapy in VTE management,^46,47 and it binds to antithrombin and thrombin (factor IIa), enhancing factor Xa inhibition, downregulating thrombin generation, thus inhibiting clot propagation. LMWH has a longer half-life and decreased risk of heparin-induced thrombocytopenia, so it can be considered superior to unfractionated heparin 48. Clinical studies and previous meta-analyses have established its efficacy in preventing VTE recurrence and a somewhat positive bleeding profile, especially in high-risk patients like those with cancer, resulting in its universal recommendation by clinical guidelines.^49-51 However, LMWH has significant limitations. Firstly, the daily subcutaneous injections put a considerable treatment burden on patients and contribute to lower adherence in the long term. Secondly, injection-site discomfort and pain reduce quality of life and further hinder sustained use. ⁵² Real-world data suggest that many patients discontinue LMWH prematurely, leaving them at risk of recurrent thrombosis. ⁵³ These challenges highlight the need for alternative drugs that not only ensure efficacy but also improve tolerability and patient convenience.

Rivaroxaban was introduced in Europe in 2008 for the prevention of venous thromboembolism (VTE), and in the United States in 2011. ⁵⁴ It is an oral, fixed-dose direct factor Xa inhibitor, and works by blocking a central step in the coagulation cascade, thereby preventing thrombin productions and clot propagation. ⁵⁵ Studies have shown that patients given rivaroxaban do not require regular monitoring, which makes management easier, and oral delivery decreases the burden of treatment and side effects associated with injectable LMWH such as pain and hematoma at the injection site. ⁵⁶ In clinical practice, the ease of administration of rivaroxaban enhances its endurance in a patient’s day-to-day life, and factor Xa inhibition offers a strong and specific anticoagulant effect. ⁵⁷ Thus, when long-term anticoagulation is needed, rivaroxaban is the drug of choice for both patients and doctors. A head-to-head comparison of rivaroxaban and LMWH on efficacy, safety, and survival outcomes is thus directly applicable to maximizing treatment regimens for VTE.

When compared to LMWH, rivaroxaban offers significant therapeutic advantages in the management of VTE. In line with results from the EINSTEIN program, where rivaroxaban showed non-inferiority to enoxaparin/VKA and equivalent protection against recurrence, we found a considerably decreased risk of total VTE recurrence with rivaroxaban. ⁵⁸ While recurrence of DVT and PE individually did not differ significantly between groups in our pooled data, these results mirror earlier RCTs that also found rivaroxaban to be equally effective to LMWH in preventing site-specific recurrences.^59,60 Interestingly, our primary analysis suggested a reduction in all-cause mortality with rivaroxaban, though this effect was attenuated in sensitivity analysis. Previous meta-analyses focusing on cancer-associated thrombosis have similarly reported mortality benefits with direct oral anticoagulants, but none specifically isolated rivaroxaban across all VTE indications.^61,62 Thus, our findings extend prior evidence by demonstrating rivaroxaban’s consistent efficacy profile across recurrence and mortality endpoints when directly compared with LMWH.

Additionally, we found no significant differences in major bleeding between rivaroxaban and LMWH. Although prior systematic reviews have largely compared direct oral anticoagulants (DOACs) with vitamin K antagonists (VKAs) rather than LMWH directly,^63,64 these broader findings similarly suggest that rivaroxaban does not confer a major excess bleeding risk relative to established anticoagulant strategies. The lack of excess gastrointestinal or intracranial bleeding parallels observations from pivotal trials such as EINSTEIN-DVT and EINSTEIN-PE, which established rivaroxaban’s non-inferior bleeding risk relative to LMWH/VKA regimens.^59,60 It should be noted that clinically relevant non-major bleeding (CRNMB) was not systematically assessed in the current review, which represents an important limitation. Previous research has suggested a potentially higher risk of CRNMB with DOACs compared to LMWH in cancer patients; however, our findings confirm that major bleeding outcomes for rivaroxaban remain comparable to LMWH in broader VTE populations. ²¹ Taken together, our results, along with prior evidence, reinforce the robustness of rivaroxaban’s efficacy and safety profile, while also highlighting our work as the most comprehensive direct comparison of rivaroxaban versus LMWH across recurrence, mortality, and bleeding outcomes.

We acknowledge that our review has several limitations. First, although we included four RCTs, the majority of the evidence base came from observational studies, which are inherently more susceptible to residual confounding and bias. Second, while checking for the risk of bias, we found “some concerns” in three of the four trials, particularly in the domain of deviations from intended interventions, which may have influenced internal validity. For observational studies, the most common limitation was incomplete control for confounders, as reflected in the comparability domain of the Newcastle–Ottawa Scale. Third, follow-up duration (ranging from 2 weeks to 24 months), baseline patient characteristics, and outcome definitions introduced heterogeneity across studies which may have affected pooled estimates. Additionally, the certainty of evidence, as assessed by the GRADE framework, showed considerable statistical difference across outcomes. While high-certainty evidence supported the effect of rivaroxaban on VTE recurrence, the evidence for DVT and PE recurrence was low certainty, and for major bleeding and all-cause mortality was very low certainty, limiting the strength of inferences for these outcomes. Finally, publication bias could not be entirely excluded, given the limited number of RCTs and potential underreporting of negative studies. Importantly, clinically relevant non-major bleeding (CRNMB) was not assessed in this review, which is a notable limitation given that prior literature has suggested a potentially higher risk of CRNMB with DOACs compared to LMWH in cancer patients; future analyses should incorporate CRNMB as a pre-specified outcome. Furthermore, while rivaroxaban and other DOACs are generally preferred for cancer-associated thrombosis, clinicians should consider clinical scenarios where LMWH may be favored, including patients with significant drug-drug interactions that affect DOAC metabolism, conditions associated with reduced gastrointestinal absorption of oral agents (e.g., malabsorption syndromes, short bowel, mucositis), and thrombocytopenia, which is frequently encountered in cancer patients on cytotoxic therapy and may limit DOAC use due to bleeding risk.

Despite these limitations, our study has several important strengths. It represents the most comprehensive synthesis to date directly comparing rivaroxaban with LMWH in patients with VTE, drawing from a large sample of over 19,000 patients. The inclusion of RCTs and high-quality observational studies provides a more comprehensive picture of treatment effects across both controlled trial settings and real-world practice. Also, the risk of bias assessments demonstrated that most observational studies had a high score on the Newcastle–Ottawa Scale, which supports the robustness of our pooled findings. In addition, the use of the GRADE framework provided a structured appraisal of evidence certainty, highlighting high-certainty evidence for reduced VTE recurrence with rivaroxaban. A further strength is that this analysis focused specifically on rivaroxaban rather than pooling it with other DOACs, allowing for a more precise evaluation of its efficacy and safety profile. Together, these strengths enhance the clinical relevance and reliability of our conclusions, providing timely evidence to guide treatment decision-making in this high-risk population.

Our findings have important implications that we need to address. Clinically, the comparable efficacy and safety of rivaroxaban compared to LMWH, combined with its practical advantages as an oral agent, suggest that it may represent a viable alternative for the long-term management of VTE. These findings guide clinician-patient shared decision-making, with potential to favor increased use of rivaroxaban in eligible candidates. From a research point of view, important gaps persist. High-certainty evidence is still lacking for key outcomes such as DVT and PE recurrence, major bleeding, and mortality. Future randomized trials with extended follow-up and incorporation of patient-reported outcomes are necessary to improve the evidence. In addition, cost-effectiveness analyses and real-world implementation studies may offer some supplemental insights as to how rivaroxaban compares to LMWH across different healthcare systems.

5. Conclusion

In this systematic review and meta-analysis of over 19,000 patients with VTE, rivaroxaban was associated with a lower risk of overall VTE recurrence compared with LMWH. In contrast, rates of DVT recurrence, PE recurrence, and major bleeding did not differ significantly between groups. A mortality benefit was observed in the primary analysis, though this was not sustained in sensitivity analyses. Altogether, our results imply that rivaroxaban offers similar safety and effectiveness to LMWH, together with the added benefits of oral administration and enhanced patient satisfaction, making it a reasonable alternative in real-world practice. Nevertheless, further high-quality trials with longer follow-up are warranted to confirm long-term outcomes.

ORCID iD

Ahmed Asad Raza https://orcid.org/0009-0004-2930-8188

Consent to Participate

This study is a systematic review and meta-analysis and did not involve human participants or collection of identifiable personal data.

Data Availability Statement

All data extracted and analyzed in this study are from published articles that are publicly accessible. The datasets used during the current study are available from the corresponding author on reasonable request.*