SAN DIEGO – Patients on rivaroxaban had significantly higher rates of GI bleeding, compared with those taking apixaban or dabigatran, results from a large population-based study showed.
“This may be due to the fact that rivaroxaban is administered as a single daily dose as opposed to the other two non–vitamin K anticoagulants [NOACs], which are given twice daily,” lead study authorsaid at the annual Digestive Disease Week. “This may lead to a greater variance in plasma drug concentration, making these patients more susceptible to bleeding.”
Mr. Ingason, a medical student at the University of Iceland, Reykjavik, said that although several studies have compared warfarin with NOACs, it remains unclear which NOAC has the most favorable GI profile. In an effort to improve the research in this area, he and his associates performed a nationwide, population-based study during March 2014–Jan. 2018 to compare the GI bleeding risk of patients receiving rivaroxaban to that of a combined pool of patients receiving either apixaban or dabigatran. They drew from the Icelandic Medicine Registry, which contains all outpatient drug prescriptions in the country. Next, the researchers linked the personal identification numbers of patients to the Landspitali University diagnoses registry, which includes more than 90% of all patients hospitalized for GI bleeding. They used 1:1 nearest neighbor propensity score for matching and Kaplan-Meier survival estimates and Cox regression to compare rates of GI bleeding. The study outcome of interest was any clinically relevant GI bleeding.
Mr. Ingason reported that the baseline characteristics were similar between the rivaroxaban group and the apixaban/dabigatran group. They matched for several variables, including age, sex, Charlson score, the proportion being anticoagulant naive, moderate to severe renal disease, moderate to severe liver disease, any prior bleeding, and any prior thrombotic events.
During the study period, 3,473 patients received rivaroxaban, 1,901 received apixaban, and 1,086 received dabigatran. After propensity score matching, the researchers compared 2,635 patients who received rivaroxaban with 2,365 patients who received either apixaban or dabigatran. They found that patients in the rivaroxaban group had significantly higher rates of GI bleeding, compared with in the apixaban/dabigatran group (1.2 and. 0.6 events per 100 patient-years, respectively). This yielded a hazard ratio of 2.02, “which means that patients receiving rivaroxaban are twice as likely to get GI bleeding compared to patients on apixaban or dabigatran,” Mr. Ingason said. When the researchers examined the entire unmatched cohort of patients, the rivaroxaban group also had significantly higher rates of GI bleeding, compared with the apixaban/dabigatran group (1.0 and 0.6 events per 100 patient-years; HR, 1.75).
Mr. Ingason and his colleagues observed that patients in the rivaroxaban group had higher rates of GI bleeding, compared with the apixaban/dabigatran group, during the entire follow-up period. At the end of year 4, the rivaroxaban group had a 4% cumulative event rate of GI bleeding, compared with 1.8% for the apixaban/dabigatran group, a highly significant difference at P = .0057).
When a meeting attendee asked Mr. Ingason why patients taking apixaban or dabigatran were combined into one group, he said that it was done to increase the power of their study. “Our theory was that rivaroxaban was different because it is administered as a single daily dose, while the others are given twice daily,” he said. The researchers reported having no financial disclosures.