RESUMEN
BACKGROUND: Liver injury (LI) has been frequently observed in patients with dilated cardiomyopathy (DCM), whereas its prognostic value remains blurry. We attempted to appraise the prognostic effect of LI in patients with DCM. METHODS: This retrospective study included 523 patients with DCM. LI was defined as a threefold increase in aspartate transaminase (≥ 135 U/L) or alanine transaminase (≥ 180 U/L) or a twofold increase in total bilirubin (≥ 41 umol/L) during hospitalization. The population was segmented into non-liver injury (NLI) group and LI group based on liver function test data. To balance differences in covariates at baseline, 1:1 propensity score matching (PSM) was performed. RESULTS: Patients with LI had lower survival rate, compared with those with NLI (44.6% vs. 73.8%, P < 0.001). Similar results were also found in age (age > 50, 39.6% vs. 70.9%, P < 0.001; age ≤ 50, 51.3% vs. 79.5%, P < 0.001) and gender stratified analysis (male, 46.2% vs. 74.4%, P < 0.001; female 35.7% vs. 72.0%, P = 0.001). After PSM, the survival rate of patients with LI remained lower than those with NLI (44.6% vs. 64.1%, P = 0.019). Multivariable Cox regression analysis manifested that LI (hazard ratio [HR]: 1.692, 95% confidence interval [CI] 1.194-2.398, P = 0.003; HR: 1.675, 95% CI 1.078-2.604, P = 0.022, respectively) showed potent predictive effect on all-cause mortality in patients with DCM, both before and after PSM. CONCLUSIONS: The occurrence of LI herald adverse outcomes in patients with DCM and attention to LI may be conducive to risk stratification and management.