Objectives: Surveillance of cirrhotic patients for early detection of hepatocellular carcinoma, based on ultrasonography and alpha1-fetoprotein determination, is a recommended practice. However, it has not been proved that this procedure can improve patient survival.
Methods: We conducted a multicenter retrospective study on 1051 consecutive patients with hepatocellular carcinoma. The criteria for eligibility were presence of underlying cirrhosis, and description of cancer stage and modalities of its diagnosis. Among 821 patients fulfilling these criteria, the tumor was detected during semiannual surveillance in 215 individuals (group 1), during annual surveillance in 155 (group 2), and as a result of symptoms or incidentally in 451 (group 3). Survival of patients under surveillance was corrected for lead time.
Results: Cancer stage was similar in groups 1 and 2 and was less advanced than in group 3 (p < 0.001). The frequency of ablative treatments or chemoembolization was similar in groups 1 and 2 and was greater than in group 3 (p < 0.001). Both surveillance programs doubled the prevalence of potential candidates for liver transplantation (68.5% and 62.5%) with respect to group 3 (32.3%, p < 0.001). However, only 15 patients underwent transplantation. In groups 1 and 2, the 5-yr survival was equivalent and was greater than in group 3 (p < 0.001). By segregating patients according to severity of cirrhosis, the benefit was confined to compensated cirrhosis (adjusted relative risk of death for patients under surveillance: 0.59 [95% CI = 0.45-0.78]).
Conclusions: Semiannual and annual surveillance equally improve the survival of cirrhotic patients with hepatocellular carcinoma and greatly increase the amenability rate to liver transplantation. When access to liver transplantation is limited, this benefit is restricted to patients with a good cirrhosis-related prognosis.