Univariate and bivariate analysis of significant risk factors for recurrent or persistent CMV infection or CMVD
| Covariate . | Relative risk (95% CI) . | |
|---|---|---|
| Univariate* . | Bivariate† . | |
| Donor CMV serology (negative) | 2.0 (1.0-3.8) | 1.6 (0.9-3.1) |
| Acute GVHD (grade 2 or higher) | 3.2 (1.4-7.0) | 2.6 (1.2-5.7) |
| Immune system recovery (absolute counts) | ||
| CD4+ | 2.2 (1.2-3.9) | 2.0 (1.2-3.4) |
| CD8+ | 2.6 (1.5-4.4) | 1.9 (1.1-3.3) |
| Tetramer-positive | 2.8 (1.3-6.3) | 2.3-2.7‡ |
| Covariate . | Relative risk (95% CI) . | |
|---|---|---|
| Univariate* . | Bivariate† . | |
| Donor CMV serology (negative) | 2.0 (1.0-3.8) | 1.6 (0.9-3.1) |
| Acute GVHD (grade 2 or higher) | 3.2 (1.4-7.0) | 2.6 (1.2-5.7) |
| Immune system recovery (absolute counts) | ||
| CD4+ | 2.2 (1.2-3.9) | 2.0 (1.2-3.4) |
| CD8+ | 2.6 (1.5-4.4) | 1.9 (1.1-3.3) |
| Tetramer-positive | 2.8 (1.3-6.3) | 2.3-2.7‡ |
CI indicates confidence interval.
All P values are less than .05.
Individual bivariate models were run to control for sample size. Tetramer-positive absolute counts remained a significant predictor when paired with each covariate. All P values were less than or equal to .05.
Tetramer relative risk range compared individually with each of the other covariates in the bivariate models.