This article is more than 5 years old. Information may no longer be current.
Early discharge, home treatment with rivaroxaban promising in low-risk pulmonary embolism
NEW ORLEANS — Certain patients with acute low-risk pulmonary embolism fared well with early discharge and home treatment with rivaroxaban, according to new data from the HoT-PE study.
“In the United States alone, around 100,000 deaths are annually attributable to PE. Although the case fatality rate appears to be decreasing over time, incident rates and overall mortality rates attributable to PE continue to increase with our aging population, showing that PE will be an increasingly greater problem for societies and health care systems in the years to come,” Stavros V. Konstantinides, MD, PhD, FESC, from the Center for Thrombosis und Hemostasis, University of Mainz, Germany, and Democritus University of Thrace, Greece, said at the American College of Cardiology Scientific Session.
Previous studies of early discharge and home treatment have been promising, he noted, but they were primarily small and the only major randomized trial was conducted 10 years ago in the era of vitamin K antagonists, according to Konstantinides. Direct oral anticoagulants, however, have simplified the initial phase of anticoagulation and transition from parenteral to oral treatments.
“The main objective of the HoT-PE trial was to determine whether early discharge and out-of-hospital treatment with rivaroxaban with patients with acute low-risk PE — with an emphasis on selected patients — would be effective, safe and, indirectly, feasible,” Konstantinides said.
Study design
Konstantinines and colleagues conducted the prospective, multicenter, international, single-arm trial with a planned enrollment of 1,050 patients with confirmed acute PE at 49 centers in seven countries throughout Europe. Because HoT-PE was a single-arm study, the researchers made assumptions about the possible outcome based on previous data, Konstantinides said. They hypothesized that the rate of the primary outcome of venous thromboembolism recurrence — symptomatic or fatal PE or deep vein thrombosis — would be less than 3% and the null hypothesis would be 3% or more. Their point estimate was 1.7% in this low-risk population, which was based on a meta-analysis of home treatment trials dating back to the vitamin K antagonist era.
After a planned interim analysis after the first 525 patients — with the objective of early termination of the trial if the null hypothesis could be rejected with 99.6% certainty — the trial stopped enrollment.
The mean age was 57 years, 46% were women and 98% were white. Most patients had good renal function, with only 5% having a creatinine clearance below 50 mL/min. Konstantinides noted that 6% of the patients had cancer and 20% had a simplified PESI score of 1 or more, suggesting that this study was more inclusive of patients for home treatment than if inclusion criteria were based on PESI or simplified PES score alone.
They used the Hestia criteria to determine if patients with PE were low-risk. The researchers also included patients who were hemodynamically stable, had no serious comorbidity that would mandate hospitalization for PE and met criteria indicating that they had the necessary social and family support to be sent home early.
Importantly, patients with right ventricular dysfunction were excluded, as data have shown that even those who are considered to be low-risk according to clinical scores have a four-fold increased risk for 30-day mortality in the presence of RV dysfunction, Konstantinides said.
Of the patients enrolled, 94% were discharged from the hospital within 48 hours, as stipulated by the protocol, and 96% spent 2 nights or less in the hospital, as tolerated by the protocol, according to Konstantinides. Nearly all patients received at least one dose of rivaroxaban (Xarelto, Janssen/Bayer) and 94% of patients completed at least 3 months of treatment.
Patients were followed for 1 year for survival.
Promising outcomes
The primary efficacy outcome — recurrence of symptomatic venous thromboembolism or fatal PE — occurred in 0.6% of the intention-to-treat population, which yielded a one-sided 99.6% upper confidence limit of 2.1%, Konstantinides said. All three events were due to recurrent PE, with no isolated DVT as an outcome and no deaths related to VTE in general.
The researchers also conducted two sensitivity analyses. In the per-protocol population, which included patients who had received at least one dose of rivaroxaban and had been discharged within 48 hours, the rate of the primary outcome was 0.4%, with a one-sided 99.6% upper confidence limit of 1.6%.
In a “worst-case scenario” analysis in which two more patients for whom incomplete data were available during follow-up were imputed as having the primary outcome, the rate was 0.95%, with a one-sided 99.6% upper confidence limit of 1.99%.
Major bleeding occurred in 1.2% of patients, including three cases of uterine bleeding, two cases of gastrointestinal bleeding and one intracranial bleed. Any clinically relevant bleeding occurred in 6% of patients and 11.2% experienced a serious adverse event.
During the 3-month follow-up, only 2.3% of the patients were rehospitalized due to suspected recurrent PE or bleeding. In the final diagnosis of these patients, there was confirmation of recurrent PE or major bleeding in 7 of 12 cases, one case of clinically relevant non-major bleeding and two cases of death — both of which were due to cancer.
“In these patients with acute low-risk PE, including absence of RV dysfunction and intracardiac thrombi, early discharge and home treatment with rivaroxaban was effective, safe and also feasible, as judged by the high rate of adherence to the protocol,” Konstantinides said. “The results of HoT-PE also support the selection of appropriate patients for ambulatory treatment with a direct oral anticoagulant, helping to reduce heart-related complications and rationalize the use of health care resources.” – by Melissa Foster
Reference:
Konstantinides S. Featured Clinical Research I. Presented at: American College of Cardiology Scientific Session; March 16-18, 2019; New Orleans.
Disclosures: Konstantinides reports he has received lecture and consultant honoraria from Actelion, Bayer AG, Biocompatibles Group UK, Boehringer Ingelheim, Daiichi Sankyo, Merck Sharpe and Dohme, Pfizer/Bristol-Myers Squibb and Servier and he has received institutional research support from Actelion, Bayer AG, Boehringer Ingelhim, Daiichi Sankyo and Merck Sharpe and Dohme.
Editor’s Note: This article was updated on March 19, 2019 to correct bleeding data. The Editors regret the error.
Perspective
Back to Top
PE and, for that matter, venous thromboembolism in general, is a major international health concern. PE is not only responsible for significant morbidity and mortality, but it is extremely costly. Physicians who care for patients with this life-threatening condition are appropriately concerned about patient safety; they worry that sending people home too early could result in untoward outcomes, such as recurrent PE or even death. As is true for PE in general, we need a more robust evidence base to guide decision-making regarding which of our patients are safe to discharge, and on what medications. In the absence of sufficient data to guide decisions, large variations in management exist. In the case of discharge planning, some physicians feeling comfortable sending patients home early, but I would say the majority do not.
The most important take-home message of this study is that, in patients with a relatively low-risk PE, it is safe to discharge early (within 48 hours) on an oral direct thrombin inhibitor, in this case rivaroxaban.
HoT-PE was a large trial that was stopped after the interim analysis showed the trial was well within its safety margin. Of course, the key to this is selecting the right patients for early discharge. The researchers used the modified Hestia criteria to help define those at risk. These criteria basically exclude patients with contraindications, such as bleeding risk. The researchers also excluded patients with RV dysfunction and/or hemodynamic instability. Utilizing these criteria, the investigators have defined a lower-risk cohort of PE patients who can now be considered safe enough to discharge early. This represents an important contribution to our knowledge base, and will definitely have widespread implications for management decisions in the future.
It is important to recognize that, just because these patients with a lower-risk PE can be sent home early on an oral anticoagulant, it does not mean that they are not at risk for PE in the future. So, long-term management and careful follow-up, even beyond what was seen in this trial, is incredibly important even in these lower-risk patients, as it is for all patients with PE.
There are several questions that remain. First, we do not know which patients should be allocated to longer-term anticoagulation therapy. Second, we do not know what the long-term consequences of a PE are on functional status. Many of us believe that impairment of functional status is underrecognized and understated, and that requires a lot of further investigation.
Overall, this trial clearly moves the field forward because it establishes a threshold below which we can feel reasonably comfortable sending a patient home on a direct oral anticoagulant. Moreover, these data will help expand the knowledge base about low-risk PE. Of course, we still need more information about higher-risk PEs and how best to manage them. The PERT Consortium’s comprehensive quality assurance database, with which Dr. Konstantinides is involved, promises to help fill that knowledge gap. Those of us who are passionate about improving treatment and outcomes in PE should all join in that initiative (www.pertconsortium.org). In the meantime, the Hot-PE trial led by Konstantinides and colleagues represents an important contribution to the literature, and will lead to better decision-making and more cost-effective care for the patient with low-risk PE.