Table 2.
The independent predictors of minor bleedings on rivaroxaban or on VKA.
| Univariate Model | Multivariate Model | |||||
|---|---|---|---|---|---|---|
| Independent Variable | p-Value | OR | 95% CI for OR | p-Value | OR | 95% CI for OR |
| MB on Rivaroxaban | ||||||
| Age, per 1 year | 0.586 | 1.010 | 0.975–1.045 | 0.996 | 1.000 | 0.929–1.076 |
| Male gender, yes/no | 0.071 | 0.454 | 0.193–1.069 | 0.828 | 0.818 | 0.134–5.005 |
| Body mass index, per 1 kg/m2 | 0.935 | 0.997 | 0.927–1.073 | 0.787 | 0.977 | 0.827–1.155 |
| Creatinine, per 1 μmol/L | 0.069 | 1.029 | 0.998–1.061 | 0.503 | 0.980 | 0.926–1.039 |
| INR, per 0.01 | 0.427 | 0.982 | 0.940–1.026 | 0.542 | 0.969 | 0.877–1.071 |
| Rivaroxaban concentration, per 1 μg/L | 0.349 | 1.011 | 0.989–1.033 | 0.947 | 1.001 | 0.970–1.034 |
| Unprovoked VTE, yes/no | 0.002 | 7.246 | 2.062–25.641 | 0.041 | 19.607 | 1.131–59.311 |
| Lag time, per 1 s | <0.001 | 1.004 | 1.002–1.006 | 0.007 | 1.006 | 1.002–1.010 |
| MB on VKA | ||||||
| Age, per 1 year | 0.811 | 1.004 | 0.969–1.042 | 0.851 | 0.996 | 0.956–1.038 |
| Male gender, yes/no | 0.594 | 1.269 | 0.528–3.051 | 0.928 | 0.948 | 0.297–3.024 |
| Body mass index, per 1 kg/m2 | 0.971 | 0.998 | 0.917–1.088 | 0.972 | 0.998 | 0.905–1.101 |
| Creatinine, per 1 μmol/L | 0.563 | 0.991 | 0.960–1.022 | 0.365 | 0.980 | 0.938–1.024 |
| INR, per 0.01 | 0.027 | 1.007 | 1.001–1.013 | 0.054 | 1.006 | 1.000–1.013 |
| Diabetes mellitus, yes/no | 0.038 | 4.386 | 1.086–17.544 | 0.059 | 4.098 | 0.948–17.857 |
| Max IIa, per 1 nM | 0.023 | 0.994 | 0.989–0.999 | 0.045 | 0.995 | 0.989–0.999 |
Abbreviations: OR: odds ratio, CI: confidence interval, MB: minor bleeding, INR: international normalized ratio, VTE: venous thromboembolism, VKA: vitamin K antagonist, lag time: time to start thrombin generation, max IIa: peak thrombin generation. For model with rivaroxaban Negelkerke R2 was 0.49 and for VKA 0.16 (p < 0.001 for both).