MARINER: No Reduction in Death, VTE With Rivaroxaban After Discharge in Medically Ill Patients

Susan Jeffrey

August 26, 2018

MUNICH — Randomized trial results show no benefit of treatment with rivaroxaban (Xarelto, Bayer/Janssen) over placebo in reducing a composite endpoint of symptomatic venous thromboembolism (VTE) or death in patients at increased risk for VTE after discharge from a medical hospitalization. 

The primary safety outcome, major bleeding, including was low and not significantly increased with treatment, the researchers said.


Results of the Medically Ill Patient Assessment of Rivaroxaban Versus Placebo in Reducing Post-Discharge Venous Thrombo-Embolism Risk (MARINER) trial were presented here at the European Society of Cardiology Congress 2018, to coincide with their publication online in the New England Journal of Medicine.

"In conclusion, our trial did not show a significant benefit of this rivaroxaban regimen started at hospital discharge with regard to the composite outcome of fatal or symptomatic venous thromboembolism in medically ill patients," the researchers, with lead author Alex C.  Spyropoulos, MD, the Donald and Barbara Zucker School of Medicine, Northwell Health at Lennox Hill Hospital, New York, write.  

However, they add, given the relatively low incidence of events despite their strategy to enrich the trial with a high-risk population, and a lack of effect seen on VTE-related death, "the usefulness of extended thromboprophylaxis remains uncertain. Future studies should more accurately identify deaths caused by thrombotic mechanisms and focus on the patients who are at highest risk and may benefit from anticoagulant prophylaxis."

During a press conference here, Spyropoulos was more optimistic in his assessment of how these results will affect practice.

"What this means from a population and other points of view, is that we now have the ability to optimize thromboprophylaxis in appropriately selected medically ill patients, and we may reduce by half or more the incidence of symptomatic venous thrombosis at the cost of little to no serious bleeding," he said.

"When we look at modelling estimations of 6 million at-risk patients that would likely qualify for MARINER inclusionary criteria, we could prevent annually up to about 10,000 fatal or nonfatal pulmonary embolic events in these patient populations annually at the cost of one-tenth that number of life-threatening or fatal bleeds."

Professor Stephan Achenbach, MD, from the University of Erlangen, Germany, chairperson of the ESC Congress Programme Committee and ESC president elect, who co-moderated the session at which MARINER was presented, said these results "will be an aid to decision making for the discharging physician to make decisions, probably not in the way that would change guidelines — that would be my current estimate."

But, he added, "since we're not making all of our decisions completely based on guidelines, but we make individual patient decisions — for example, the patient who is at particularly high risk for VTE and particularly low risk for bleeding — probably a physician who really takes care would take this into consideration."

Spyropoulos was asked during the press conference, though, how they could make that conclusion on the basis of a neutral primary composite endpoint in MARINER.

"What we saw in MARINER was that there were two components of the primary endpoint, death due to venous thrombosis component, and symptomatic venous thrombosis component," Spyropoulos responded. "So, where we were not successful was in the first part of that component, but we were quite successful in the second part.

"So as our moderator stated, I think it then becomes up to the individual physician at the bedside to say, given these results in the high-risk patient, do I want to prevent symptomatic nonfatal pulmonary embolic events at the cost of little or no bleeding? And my interpretation of those results is, if the patient is at high enough risk, as defined by the IMPROVE score or an elevated D-dimer, then I think the answer would be yes."

Ileana L. Piña, MD, MPH, a heart failure specialist at Montefiore Medical Center, Bronx, New York, is not convinced. She agreed that the period following a hospitalization is a "very perilous time," where patients tend to be immobile, are learning to take new medications, and are likely not to be in cardiac rehabilitation immediately, "so you worry about that time period."

In this trial, rivaroxaban was "pretty safe and less bleeding probably than aspirin, but there was really no benefit, and pretty much no benefit in anything. You can tear it up and say it was the dosing, but it was neutral," Piña said. 

Events seemed to be less severe when they did occur, she added, "but I think there's a lot of confounders in there that you just don't know about. And it's expensive…so I don't know that it's worth the expense at this point.

"I've never done it on my patients, and I have heart failure patients who are sick, and who are not very mobile, and we don't do that." 


Patients hospitalized for acute medical illnesses are at increased risk for VTE events, the authors note in their report, and risk scores that include factors such as lack of mobility, advanced age, cancer, previous VTE, and elevated D-dimer levels help identify those at risk for symptomatic VTE.

Anticoagulant prophylaxis while patients are in the hospital has been shown to reduce the risk for VTE by 50% to 60%, they write, but prophylaxis is "rarely continued after discharge in accordance with current guidelines," despite the continued risk for symptomatic VTE — including fatal pulmonary embolism — for 6 weeks or more after discharge.

"However, studies of extended thromboprophylaxis have shown either excess major bleeding or a benefit that is based mainly on reducing the risk of asymptomatic deep-vein thrombosis," the investigators write.

"As such, the MARINER trial was designed to optimize the benefit-risk profile of extended prophylaxis with the direct oral anticoagulant rivaroxaban at hospital discharge in an at-risk medically ill population, using clinically meaningful, and guideline-recommended endpoints," Spyropoulos said during the presentation here. "We thus used both a VTE-enrichment strategy as well as reduced dosing in subjects with moderate renal sufficiency and added key exclusionary criteria to improve safety."

The trial randomly assigned 12,024 patients at least 40 years of age who had been hospitalized for an acute illness for at least 3 but not more than 10 consecutive days to receive the direct oral anticoagulant, at 10 mg once daily for patients with a creatinine clearance of at least 50 mL per minute, 7.5 mg once daily for those with creatinine clearance at least 30 but less than 50 mL per minute, or a placebo for 45 days.

"This duration was chosen because previous studies had shown that approximately 75% of post-hospital discharge venous thromboembolic events occur by 45 days after discharge," the authors note.

Conditions underlying the hospitalizations included heart failure with a left ventricular ejection fraction of 45% or less, acute respiratory insufficiency or exacerbation of chronic obstructive pulmonary disease, acute ischemic stroke, or acute infectious or inflammatory disease, including rheumatic diseases.

Eligible patients were assessed to be at increased risk for VTE on the basis of a modified International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) score of 4 or more, or a score of 2 or 3 plus a plasma D-dimer level of more than twice the upper limit of the normal range, defined according to local laboratory criteria. They also had to have received thromboprophylaxis during the index hospitalization with low-molecular-weight or unfractionated heparin.

Patients were excluded if they had any condition that required anticoagulant or dual antiplatelet therapy, if they had active cancer, if they'd had bleeding within 3 months or a had a high risk of bleeding, or if they had any other contraindication to rivaroxaban.

The primary efficacy outcome, a composite of any symptomatic venous thromboembolism including deep-vein thrombosis in the legs or nonfatal pulmonary embolism, or death related to venous thromboembolism (ie, death from pulmonary embolism or death in which pulmonary embolism could not be ruled out as the cause).

Results for the primary efficacy endpoint showed no significant difference between the groups overall or by creatinine clearance; the difference in risk for rivaroxaban minus placebo was –0.27 percentage points (95% confidence interval [CI], –0.61 to 0.08).

Table 1. MARINER: Primary Efficacy Endpoint

Endpoint Rivaroxaban, n (%) Placebo, n (%) Hazard Ratio (95% CI) P Value
Symptomatic VTE or death due to VTE        
  Overall 50/6007 (0.83) 66/6012 (1.10) 0.76 (0.52 - 1.09) .14
Creatine clearance ≥ 50 mL/min, 10-mg dose 32/4909 (0.65) 48/4913 (0.98) 0.67 (0.43 - 1.04) .075
Creatine clearance <50 mL/min, 7.5-mg dose 18/1098 (1.64) 18/1099 (1.64) 1.00 (0.52 - 1.92) .994

When the primary endpoint results were parsed by dose stratum and baseline renal function, they found a higher incidence with moderate renal insufficiency at 1.64% and no effect of the reduced dose on this risk. "Conversely, for the rest of the trial population, the incidence of the primary efficacy outcome was 0.98% in the placebo group versus 0.65% in the rivaroxaban 10 mg group, suggesting a treatment effect."

If superiority for rivaroxaban over placebo had been shown, then secondary outcomes were to have been tested sequentially with the use of the same Cox proportional hazards model in the hierarchical order shown below, they note. However, because superiority was not established in the primary analysis, they write, "the prespecified secondary efficacy outcomes were assessed as exploratory analyses without adjustment for multiplicity."

"Although rivaroxaban had no effect on the risk of venous thromboembolism-related death, it was associated with fewer symptomatic venous thromboembolic events than placebo (risk difference, –0.24 percentage points)." The exploratory analysis showed a 56% reduction in symptomatic VTE and a 27% reduction in VTE or all-cause mortality, he noted, although neither finding reached statistical significance.

Table 2. MARINER: Secondary Efficacy Endpoints

Endpoint Rivaroxaban (%) Placebo (%) Hazard Ratio (95% CI) P Value
VTE-related death 0.72 0.77 0.93 (0.62 - 1.42) .751
Symptomatic nonfatal VTE 0.18 0.42 0.44 (0.22 - 0.89) .023
Symptomatic VTE or death from any cause 1.30 1.78 0.73 (0.54 - 0.97) .033
Symptomatic VTE, myocardial infarction, nonhemorrhagic stroke, cardiovascular death 1.56 2.00 0.78 (0.60 - 1.02) .073
Death from any cause 1.18 1.48 0.80 (0.58 - 1.09) .156

Major bleeding, the primary safety endpoint, was defined as overt bleeding associated with a decrease in the hemoglobin level of 2 g/dL or more, bleeding that required transfusion of 2 or more units of packed red cells or whole blood, bleeding that occurred at a critical site, or fatal bleeding.

The difference between groups in terms of major bleeding was "low," the authors said, at 0.13 percentage points.

Table 3. MARINER: Primary Safety Endpoint

Endpoint Rivaroxaban Placebo Hazard Ratio (95% CI) P Value
Major bleeding n (%) 17/5982 (0.28) 9/5980 (0.15) 1.88 (0.84 - 4.23) .124


"These differences in risk suggest that the number of patients needed to treat to prevent one symptomatic venous thromboembolism event is 430, whereas the number needed to cause one major bleed is 856," they write.

"Thus, although the benefit-risk decision for the individual patient is finely tuned, the implementation of extended thromboprophylaxis with appropriate selection of medically ill patients may reduce the health burden of nonfatal venous thromboembolism in this population."

Subgroups May Still Benefit?

Discussant for the paper at the congress here was Malte Kelm, MD, University of Düsseldorf, Germany. He concluded that "MARINER suggests that direct oral anticoagulants should not be prescribed post-discharge to medically ill patients in general, but specific subgroups may profit."

For example, he pointed to patients with heart failure, who constituted about 40% of patients in the trial, "which is not surprising," since these patients carry higher VTE risk for many reasons and have recently been shown in the MAGELLAN trial that VTE events increase with increasing severity of heart failure.

The trial also supports the concept of appropriate risk assessment for VTE in medically ill patients, he said, and "offers the perspective of tailored indication and dosing for direct oral anticoagulants in specific subgroups of patients, depending on comorbidities, and implicates that rivaroxaban with prophylactic dose provides sufficient safety in extended VTE prophylaxis."

During the presentation, audience members participated in a before-and-after poll, asked first what regimen of VTE prophylaxis they use after hospital discharge of medically ill patients at high risk for pulmonary embolism. The most popular answer prior to the presentation showed 41% reported low-dose low-molecular-weight heparin during the admission, with others reporting they use oral anticoagulants, or only treat in very high risk patients.

Afterward, the most common response among poll participants was that the study findings would not change their current practice, at 46%, with 27% saying they felt these findings were not conclusive.

MARINER was supported by Janssen Research and Development. Spyropoulos reports grants and personal fees from Janssen during the conduct of the study; personal fees from Daiichi Sankyo, grants and personal fees from Boehringer Ingelheim, personal fees from Portola, personal fees from Bayer, personal fees from ATLAS (Colorado Prevention Center), outside the submitted work.

European Society of Cardiology (ESC) Congress 2018. Hotline Session 1: #2066. Presented August 26, 2018

N Engl J Med.  Published online August 26, 2018. Full text

For more from | Medscape Cardiology, follow us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: