Table 3.
Variables associated with composite graft loss in the study cohort.
| Variables | Univariable analyses | Multivariable analyses | ||
|---|---|---|---|---|
| HR (95% CI) | P | aAdjusted HR (95% CI) | P | |
| Increased post-TPL RDW (>14.9%) | 2.03 (1.60–2.58) | <0.001 | 1.60 (1.23–2.07) | <0.001 |
| Age (1-year increment) | 1.02 (1.01–1.03) | <0.001 | 1.00 (0.99–1.01) | 0.76 |
| Male sex (vs. female) | 1.37 (1.08–1.74) | 0.01 | 1.26 (0.95–1.66) | 0.11 |
| Body mass index (1 kg/m2 increment) | 1.06 (1.02–1.09) | 0.003 | 1.24 (0.98–1.07) | 0.26 |
| Smoking (vs. never) | 1.29 (0.98–1.69) | 0.07 | 1.14 (0.85–1.54) | 0.38 |
| Diabetes mellitus (vs. none) | 1.94 (1.46–2.57) | <0.001 | 1.74 (1.28–2.38) | <0.001 |
| Post-TPL 3 month eGFR < 30 mL/min/1.73 m2 (vs. eGFR ≥ 30 mL/min/1.73 m2) | 1.78 (1.22–2.61) | 0.003 | 1.37 (0.91–2.08) | 0.13 |
| Underlying primary glomerulopathy (.vs other cause of ESRD) | 0.69 (0.50–0.95) | 0.02 | 0.83 (0.60–1.16) | 0.28 |
| Hemoglobin (1 g/dL increment) | 0.89 (0.83–0.95) | <0.001 | 0.93 (0.87–1.00) | 0.06 |
| Hypoalbuminemia (<3.0 g/dL) | 2.12 (1.13–3.98) | 0.02 | 1.36 (0.71–2.59) | 0.35 |
| Deceased donor (.vs other donor relationship) | 1.63 (1.27–2.08) | <0.001 | 1.56 (1.20–2.03) | <0.001 |
| Number of mismatched HLA (one antigen increment) | 1.00 (0.79–1.27) | 0.99 | 1.07 (0.99–1.16) | 0.09 |
| Acute rejection within 3 months (vs. none) | 1.50 (1.05–2.16) | 0.03 | 1.20 (0.83–1.75) | 0.34 |
| Induction therapy (vs. none) | 0.92 (0.71–1.19) | 0.51 | 0.72 (0.54–0.96) | 0.02 |
| Use of tacrolimus (vs. no use) | 1.00 (0.79–1.27) | 0.99 | 1.03 (0.80–1.33) | 0.82 |
HR, hazard ratio, CI, confidence interval, DM, diabetes mellitus, eGFR, estimated glomerular filtration rate, ESRD, end stage renal disease, HLA, human leukocyte antigen, RDW, red cell distribution width, TPL, transplantation. Missing data was imputed by multiple imputation by classification and regression trees (CART) method, and hazard ratio and associated confidence of each characteristic was shown. aThe multivariable cox regression model was adjusted for all variables in the Table 3, which were the baseline characteristics which were significantly different according to presence of increment RDW.