Table 4.
Multivariate logistic regression analysis; predictors of BK viremia
| Variable | β | S.E. | OR | 95% CI | p-value |
|---|---|---|---|---|---|
| Study participant (yes vs. no) |
−1.82 |
0.56 |
0.16 |
0.054 – 0.49 |
0.001 |
| BPAR (yes vs. no) |
1.33 |
0.47 |
3.79 |
1.50 – 9.58 |
0.005 |
| Donor CMV IgG negative (yes vs. no) |
1.00 |
0.45 |
2.71 |
1.13 – 6.51 |
0.026 |
| Recipient CMV IgG positive (yes vs. no) |
1.05 |
0.49 |
2.84 |
1.10 – 7.37 |
0.031 |
| Previous transplantation (vs. first) | 1.009 | 0.49 | 2.74 | 1.05 – 7.15 | 0.039 |
For multivariate analyses all parameters that were significant in the univariate logistic regression model as well as the following known risk factors for BK viremia that were negative in the univariate approach were entered into the model: recipient as well as donor sex, donor age, use of CNI (tacrolimus or CyA), ATG, HLA mismatch).