Table 4.
Predictors of development of de novo post-transplant type 2 diabetes (adjusted hazard ratio with 95 % confidence interval is calculated using Cox proportional hazard model). Total N used in the model = 13,000: nc = 9999 censored, ne = 3,001 events (de novo DM onset)
| Predictor | aHR (95 % CI) | p |
|---|---|---|
| NASH | 1.29 (1.18–1.41) | <.0001 |
| Calendar year | 0.92 (0.91–0.94) | <.0001 |
| Age at transplant, per year | 1.02 (1.01–1.02) | <.0001 |
| African-American | 1.32 (1.16–1.51) | <.0001 |
| Hispanic | 1.11 (0.99–1.24) | 0.08 |
| Asian | 0.99 (0.84–1.17) | 0.92 |
| Male gender | 1.11 (1.03–1.19) | 0.0080 |
| Overweight | 1.13 (1.04–1.24) | 0.0061 |
| Obese | 1.32 (1.21–1.45) | <.0001 |
| Liver malignancy | 0.96 (0.86–1.08) | 0.49 |
| Pre-transplant CAD | 1.19 (0.92–1.53) | 0.19 |
| Pre-transplant hypertension | 1.05 (0.96–1.15) | 0.27 |
| Donor’s age, per year | 1.002 (1.000–1.004) | 0.0414 |
| Procurement from a non-heart-beating donor | 1.24 (1.04–1.46) | 0.0140 |
| Donor’s history of DM | 1.09 (0.96–1.22) | 0.18 |
| Use of tacrolimus *) | 0.98 (0.83–1.15) | 0.80 |
| Use of mycophenolates *) | 0.92 (0.84–1.02) | 0.11 |
| Use of steroids *) | 1.87 (1.49–2.34) | <.0001 |
*) Ever used before the first onset of post-transplant DM