Safety of direct oral anticoagulants in patients with hereditary hemorrhagic telangiectasia

Table 3 Relationship between Rivaroxaban and bleed severity calculated by multiple regression

Adjusted for	Coefficient for Rivaroxaban (95% confidence intervals)	Adjusted r squared for model	P value
Nil (crude regression)	0.67 (0.04, 1.30)	0.14	0.038
Sex	0.67 (0.024, 1.31)	0.17	0.043
Age	0.74 (0.12, 1.36)	0.18	0.021
Indication (AF or VTE)	0.79 (0.18, 1.41)	0.22	0.014
Age and indication	0.81 (0.18, 1.43)	0.22	0.014
Age, sex and indication	0.89 (0.26, 1.51)	0.24	0.008

Linear regression of Rivaroxaban against outcome variable of the 4 point nosebleed scale (improved, unchanged, worse but tolerated, worse leading to treatment discontinuation) across the 25 patients with a single treatment episode, compared to the other two agents (11 Rivaroxaban-users vs. 14 non-Rivaroxaban-users, either Apixaban or Dabigatran). There were similar relationships when including the three patients who had used more than one DOAC (data not shown)

ISSN: 1750-1172