Table 2.
Adjusted Logistic Models of Clinically Relevant DDI as the Dependent Variable for Each DOAC.a
| Rivaroxaban | Apixaban | Edoxaban | Dabigatran | |||||
|---|---|---|---|---|---|---|---|---|
| Adjusted OR (95% CI) | P value | Adjusted OR (95% CI) | P value | Adjusted OR (95% CI) | P value | Adjusted OR (95% CI) | P value | |
| CYP3A4 substrate | 3.02 (0.59-15.42) | .183 | 3.56 (0.82-15.44) | .091 | 3.36 (0.68-16.64) | .138 | 3.36 (0.68-16.64) | .138 |
| P-glycoprotein substrate | 0.82 (0.26-2.60) | .729 | 1.17 (0.42-3.26) | .771 | 0.69 (0.20-2.33) | .544 | 0.69 (0.20-2.33) | .544 |
| CYP3A4 competition | 0.08 (0.01-0.58) | .013 | 0.08 (0.01-0.61) | .015 | 0.09 (0.01-0.99) | .049 | 0.09 (0.01-0.99) | .049 |
| P-glycoprotein competition | 5.02 (0.19-133.75) | .335 | 0.01 (0.0-962.8) | .348 | 0.13 (0.01-8.38) | .337 | 0.13 (0.01-8.38) | .337 |
| CYP3A4 inhibition | 1.84 (0.75-4.52) | .186 | 2.19 (1.03-4.63) | .041 | 1.56 (0.63-3.84) | .334 | 1.56 (0.63-3.84) | .334 |
| CYP3A4 induction | 5.22 (2.17-12.54) | <.001 | 4.63 (2.03-10.57) | <.001 | 1.37 (0.52-3.65) | .528 | 1.37 (0.52-3.65) | .528 |
| P-glycoprotein inhibition | 2.14 (1.17-3.91) | .014 | 2.05 (1.17-3.59) | .013 | 4.87 (2.17-10.93) | <.001 | 4.87 (2.17-10.93) | <.001 |
| P-glycoprotein induction | 0.46 (0.09-2.31) | .346 | 4.95 (0.06-426.65) | .482 | 18.56 (0.85-404.76) | .063 | 18.56 (0.85-404.76) | .063 |
| Tyrosine kinase inhibitors | 6.60 (2.08-20.95) | .001 | 4.63 (2.03-10.57) | <.001 | 8.67 (2.65-28.38) | <.001 | 8.67 (2.65-28.38) | <.001 |
Abbreviations: AIC, Akaike; BIC, Bayesian; DDI, drug–drug interactions; DOAC, non-vitamin K antagonist anticoagulants; OR, odds ratio.
a Models: rivaroxaban: BIC = 139.7, AIC = 104.2, C-statistics 0.91 (95% CI: 0.85-0.97), P < .001; apixaban: BIC =138.4, AIC = 102.9, C-statistics 0.86 (95% CI: 0.78-0.93), P < .001; edoxaban: BIC = 125.41, AIC = 89.9, C-statistics 0.93 (95% CI: 0.88-0.98), P < .001; dabigatran: BIC = 125.41, AIC = 89.9, C-statistics 0.93 (95% CI: 0.88-0.98), P < .001.