← Back
NeuroTrials.ai
Neurology Clinical Trial Database

TRACK

Low-Dose Rivaroxaban and Cardiovascular Events in Advanced Kidney Disease: The TRACK Randomized Clinical Trial

Share Share Copied! Compare

Year of Publication: 2026

Authors: Badve SV, Perkovic V, Jha V, ..., Gallagher M; for the TRACK Trial Investigators

Journal: JAMA

Citation: JAMA. 2026;336(4):296-305. doi:10.1001/jama.2026.9379

Link: https://doi.org/10.1001/jama.2026.9379

Clinical Question

Does low-dose rivaroxaban (2.5 mg twice daily) reduce major adverse cardiovascular events compared with placebo in patients with advanced chronic kidney disease?

Bottom Line

In patients with advanced CKD (stage 4/5 or dialysis-dependent) at high cardiovascular risk, low-dose rivaroxaban did NOT reduce cardiovascular events and significantly increased major bleeding. Low-dose rivaroxaban should not be used for cardiovascular event prevention in this population.

Major Points

  • First large RCT of low-dose rivaroxaban in advanced CKD (stage 4/5 including dialysis)
  • No cardiovascular benefit: primary composite outcome 22.6% vs 20.7% (HR 1.09, P=.46)
  • Significant harm: major bleeding 8.8% vs 6.0% (HR 1.51, 95% CI 1.02-2.22, P=.04)
  • Trial stopped early on August 7, 2025 for lack of efficacy on DSMB recommendation
  • Findings do NOT support extending COMPASS-style low-dose rivaroxaban regimen to advanced CKD patients

Design

Study Type: Randomized, double-blind, placebo-controlled clinical trial

Randomization: 1

Blinding: Double-blind (participants, investigators, study coordinators, site pharmacists, treating physicians, outcome assessors all blinded)

Allocation: 1:1 via password-protected web-based system using covariate-balancing adaptive allocation algorithm minimizing imbalances by study site, aspirin use at randomization, CKD stage, and diabetes status

Enrollment Period: January 18, 2021 to July 31, 2025

Follow-up Duration: Median 1.7 years; final follow-up October 30, 2025

Centers: 90

Countries: Australia, Belgium, Canada, France, Germany, India, Malaysia, Nepal, Saudi Arabia, Singapore, Taiwan, Tunisia

Sample Size: 1458

Analyzed: 1458

Analysis: Intention-to-treat; shared-frailty Cox model with random site effect, including 3 stratification variables (baseline aspirin use, CKD stage, diabetes) as fixed covariates. Final significance level set at 2-sided 4.86% accounting for 1 interim analysis. Holm-Sidak step-down for 5 secondary outcomes.

Power Calculation: Original plan: 1800 participants to detect 25% risk reduction (HR 0.75), assuming placebo event rate 10/100 person-years, 90% power, 2-sided alpha 5%; required 515 primary events (also 80% power for 22% reduction, HR 0.78). Trial stopped early at 1458 patients for lack of efficacy.

Registration: ClinicalTrials.gov NCT03969953

Inclusion Criteria

  • Adults ≥18 years
  • Dialysis-dependent kidney failure OR CKD stage 4/5 not on kidney replacement therapy (eGFR ≤25 mL/min/1.73 m²)
  • Increased cardiovascular risk defined by ≥1 of: coronary artery disease, nonhemorrhagic/nonlacunar stroke, peripheral artery disease, diabetes, or age ≥65 years
  • Successful completion of run-in phase (≥80% adherence to placebo over 21 days [range 14-30])

Exclusion Criteria

  • Mechanical or prosthetic heart valve (other than bioprosthetic)
  • Indication for or contraindication to anticoagulant therapy
  • High bleeding risk per treating clinician
  • Lesion/condition significantly associated with major bleeding risk
  • Major bleeding within past 30 days or active major bleeding
  • History of hemorrhagic or lacunar stroke; any stroke within past month
  • Treatment with P2Y12 inhibitor or phosphodiesterase inhibitor that cannot be discontinued
  • Concurrent strong inhibitors of combined CYP3A4 and P-glycoprotein or strong CYP3A4 inducers
  • Severe heart failure (LVEF <30% or NYHA class III/IV)
  • Uncontrolled hypertension at screening (SBP >180 or DBP >110 mm Hg)
  • Hemoglobin <90 g/L, platelet count <100 × 10⁹/L
  • Significant liver disease (Child-Pugh B or C), ALT >3× ULN
  • Kidney transplant with functioning allograft or scheduled living-donor transplant
  • Pregnancy or intention to become pregnant or breastfeed (premenopausal women excluded in Europe)
  • Inability to understand or comply with study requirements
  • Atrial fibrillation (unless therapeutic anticoagulation not indicated)

Baseline Characteristics

Overall Mean Age (SD): 63.2 (11.6) years

Female: 432 (29.6%)

Total Randomized: 1458

Rivaroxaban Arm N: 727

Placebo Arm N: 731

Follow-up Completion: 1360 (93.3%)

Arms

FieldLow-dose rivaroxabanControl
N727731
InterventionRivaroxaban 2.5 mg twice dailyMatching placebo twice daily
DurationMedian 1.7 years (trial stopped early)Median 1.7 years (trial stopped early)

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Composite of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or nonfatal peripheral artery disease eventPrimary151/731 (20.7%); 11.8 events per 100 person-years164/727 (22.6%); 13.0 events per 100 person-years1.090.46
All-cause mortalitySecondaryNot reported in abstract
Individual components of primary composite (CV death, nonfatal MI, nonfatal stroke, nonfatal PAD event)SecondaryNot reported in abstract
Composite of CV death, nonfatal MI, or strokeSecondaryNot reported in abstract
Composite of all-cause mortality, nonfatal MI, or strokeSecondaryNot reported in abstract
Composite of all-cause mortality, nonfatal MI, stroke, or PAD eventSecondaryNot reported in abstract
Venous thromboembolismSecondaryNot reported in abstract
Major bleeding (composite of fatal bleeding, symptomatic bleeding into critical area/organ, bleeding into surgical site requiring reoperation, or bleeding leading to hospitalization)Safety44/731 (6.0%); 3.4 events per 100 person-years64/727 (8.8%); 5.1 events per 100 person-years1.510.04
Gastrointestinal bleedingSafetyPrespecified safety outcome; specific numbers not reported in extracted text

Subgroup Analysis

Prespecified subgroup analyses examined effect modification; specific results not provided in extracted text. A specific analysis was planned to evaluate whether bleeding rates were higher among Asian participants.

Criticisms

  • Trial stopped early for lack of efficacy at 1458 of planned 1800 patients, which may limit precision for secondary and subgroup analyses
  • Median follow-up of only 1.7 years may be insufficient to detect long-term cardiovascular benefits
  • Aspirin use at baseline was a stratification variable but the combination effect (vs COMPASS regimen of rivaroxaban + aspirin) is not clearly separable in this abstract
  • Broad inclusion criteria (age ≥65 alone qualifies) may dilute treatment effect in truly high-risk patients

Funding

Not specified in extracted text

Based on: TRACK (JAMA, 2026)

Authors: Badve SV, Perkovic V, Jha V, ..., Gallagher M; for the TRACK Trial Investigators

Citation: JAMA. 2026;336(4):296-305. doi:10.1001/jama.2026.9379

Content summarized and formatted by NeuroTrials.ai.