70 Multicenter Trial of Rivaroxaban for Early Discharge of Pulmonary Embolism from the Emergency Department

Abstract

Hospitalization after an ED diagnosis of a pulmonary embolism (PE) is common, expensive, and of questionable clinical benefit in low-risk patients. Our purpose was to determine if low-risk PE patients, discharged home from the ED on rivaroxaban, have a different total number of hospital days through day 30 vs standard of care (SOC). This was a multicenter, prospective, open label, randomized clinical trial of patients ≥18 years of age with an ED diagnosis of low risk PE (defined by HESTIA criteria). After consent subjects were randomized by a central computer to SOC, which was determined by the ED attending, or ED discharge on rivaroxaban, and followed for 90 days. Our primary outcome was the total number of initial hospital days, plus days of hospitalization related to bleeding or VTE, during the 30 days after a low risk ED PE diagnosis in patients discharged on rivaroxaban vs SOC. A 90-day composite safety endpoint was defined as International Society of Thrombosis and Hemostasis (ISTH) major bleeding, clinically relevant non-major bleeding, and mortality. Cohorts were compared using descriptive statistics and 95% confidence intervals for mean differences. We randomized 114 subjects, 51 to rivaroxaban and 63 SOC. Most were white (67.5%), female (51.8%), with a median (IQR) age of 50 (36, 61) yrs. Of 112 (98.2%) receiving at least 1 dose of study drug, 99 (86.8%) completed the study. There were 15 (13.2%) discontinuations; 7 (6.1%) lost to follow-up, 4 (3.5%) adverse events, and 2 (1.8%) consent withdrawal. Median (IQR) therapy duration was 91 (89, 94), and 89 (85, 93) days for rivaroxaban vs SOC. The intent to treat primary endpoint was 4.6 hours for rivaroxaban vs 33.4 hours for SOC (mean difference -1.20 days, 95% CI -1.773, -0.630). Mean total hospital days (for any reason) at 90 days after randomization were significantly shorter for rivaroxaban than SOC; 0.8 vs. 1.8 days (mean difference between groups was -0.8 days; 95% CI: -0.963, -0.605). At 90 days follow-up there were no ISTH bleeding events, recurrent VTE, or deaths. Unplanned VTE and bleeding related hospital or doctor visits within 90 days were numerically lower with rivaroxaban; 2 (3.9%) vs. 4 (6.3%); difference in proportions -0.02 (95% CI: -0.21 to 0.16). The composite safety endpoint was similar in both groups; difference in proportions 0.005 (95% CI: -0.181 to 0.191). There were numerically more adverse events (AEs) with rivaroxaban than SOC, 59.2% vs. 39.7%, however, the overall rate of serious AEs, AEs leading to hospitalization, and AEs leading to study drug discontinuation were similar between cohorts (p>0.05). In this prospective, open label, randomized standard-therapy controlled trial, low risk ED PE patients discharged on rivaroxaban had similar outcomes as SOC but spent fewer total days in the hospital in the month following their ED discharge.