Atrial fibrillation in patients with cirrhosis

Few information is available regarding atrial fibrillation in cirrhotic patients. The aim of this study was to investigate the occurrence and clinical impact of atrial fibrillation in these patients.

Three hundred and thirty-five cirrhotic patients (219 males; mean age 65 ± 10.85 years; 196 Child-Pugh class A, 104 class B and 35 class C) were consecutively analysed and followed up for 24 months. Electrocardiograms were available for all patients before starting the study, at basaltime and during the follow-up. Echocardiography was performed in individuals with atrial fibrillation and in 100 randomly chosen patients without it.

Atrial fibrillation was observed in 21/335 cirrhotics (mean age 75 ± 7 years, 13 male), six of whom had permanent and 15 had paroxysmal atrial fibrillation. At univariate analysis, atrial fibrillation significantly correlated with older age, history of coronary heart disease, Child-Pugh score, serum albumin, hepatic encephalopathy, treatment with furosemide, QTc prolongation, atrial section areas, increased PAPs and thickness of interventricular septum. Age [odd ratio 1.12, 95% CI (1.05-1.2), P = 0.001], history of coronary heart disease [odd ratio 4.93, 95% CI (1.04-23.54), P = 0.04] and PAPs [odd ratio 1.12, 95% CI (1.02-1.2), P = 0.01] maintained statistical significance at multivariate analysis. Fifty-one of the 335 patients died during the follow-up. At Cox regression analysis, advanced Child-Pugh score [hazard ratio 1.546, 95% CI (1.357-1.762), P = 0.037] and increased heart rate [hazard ratio 1.117, 95% CI (1.021-1.223), P = 0.016] were significantly associated with mortality which was independent of atrial fibrillation occurrence.

Cirrhosis is not a predisposing factor of atrial fibrillation, which in turn has no impact on mortality in cirrhotic patients.