Prolonged Q-T interval predicts severe arrhythmias and sudden death, and has been shown to occur in alcoholic liver disease and cirrhotic patients who are candidates for liver transplantation. This study first evaluated the prevalence of prolonged Q-T interval in a large population of unselected patients with cirrhosis, and assessed the relationship between abnormal Q-T, etiology, and severity of liver disease and mortality of patients. Possible causes of Q-T abnormality were also explored. Ninety-four patients with cirrhosis without overt heart disease and 37 control subjects with mild chronic active hepatitis were enrolled. Rate-corrected Q-T interval (Q-Tc) was assessed along with routine liver tests, Child-Pugh score, serum bile salts, electrolytes and creatinine, plasma renin activity, aldosterone, norepinephrine, atrial natriuretic factor and, gonadal hormones. Q-Tc was longer in patients with cirrhosis than in controls (440.3 +/- 3.2 vs. 393.6 +/- 3.7 ms; P < .001) and prolonged (> 440 ms) in 44 patients (46.8%) and 2 controls (5.4%; P < .001). Q-Tc length was not influenced by the etiology of cirrhosis and correlated with Child-Pugh score (r = .53; P < .001), liver tests such as prothrombin activity, and serum concentrations of albumin and bilirubin, plasma bile salts, and plasma norepinephrine. Multivariate analysis showed that only Child-Pugh score and plasma norepinephrine were independently correlated with Q-Tc duration. Over a median follow-up period of 19 months (range, 2-33 months), patients with Q-Tc longer than 440 ms had a significantly lower survival rate than those with normal Q-Tc. Q-T interval is frequently prolonged in patients with cirrhosis, regardless the etiology of the disease, worsens in parallel with the severity of the disease, and may have an important prognostic meaning. In addition to other undefined factors related to the severity of cirrhosis, sympathoadrenergic hyperactivity may play a pathogenetic role.