Multivariate Cox regression analysis of variables associated with death after listing for liver transplantation in PRS-matched patients with refractory ascites
| Cox regression analysis | Competing risk Cox regression analysis | |||||
|---|---|---|---|---|---|---|
| Outcome death (censored at transplant) | Outcome death | |||||
| HR | 95% CI | p Value | HR | 95% CI | p Value | |
| MELD score | 1.07 | 0.97 to 1.19 | 0.197 | 0.98 | 0.90 to 1.06 | 0.600 |
| Hyponatraemia | 2.70 | 0.84 to 8.67 | 0.095 | 1.32 | 0.50 to 3.52 | 0.579 |
| Previous variceal haemorrhage | 0.69 | 0.18 to 2.64 | 0.590 | 0.45 | 0.12 to 1.65 | 0.229 |
| Prophylactic antibiotics | 0.60 | 0.19 to 1.95 | 0.395 | 0.39 | 0.14 to 1.08 | 0.070 |
| NSBB | 0.33 | 0.12 to 0.89 | 0.028 | 0.35 | 0.14 to 0.86 | 0.022 |
p Value in bold indicates statistical significance.
In the competing risk Cox regression analysis, the only variable associated with the outcome transplant was MELD (adjusted HR 1.12; 95% CI 1.04 to 1.20, p=0.002) (model not shown). Hepatocellular carcinoma excluded from the analysis because of small number of patients (n=3).
Reference group (relative risk 1.00): no hyponatraemia, no previous variceal haemorrhage, no prophylactic antibiotics and no NSBB.
MELD, model for end-stage liver disease; NSBB, non-selective β-blockers; PRS, propensity risk score.