Utilization of screening for hepatocellular carcinoma in the United States

BACKGROUND: Screening for hepatocellular carcinoma (HCC) has been recommended for patients at high-risk of developing HCC. Yet, the utilization and determinants of screening remain unclear. METHODS: All patients diagnosed with HCC at 3 medical centers during 1998 to 2003 were identified. Information regarding receipt of HCC screening, demographics, risk factors, liver disease severity, number of HCC lesions, therapy, and date of death was abstracted from medical records. Multivariable logistic regression models were conducted to evaluate determinants of HCC screening and therapy. Cox proportional hazards models were developed to assess the effect of screening on risk of mortality. RESULTS: We identified 157 patients diagnosed with HCC. The majority of patients were <65 years (62%), white (59%), had a single mass (42%), and a Child-Pugh-Turcotte score B (41%). Approximately, 28% (n=44) received at least one possible screening test (36% alpha-fetoprotein only, 23% abdominal ultrasound only, 7% computed tomography only; 34% had more than one test). Screened patients were younger [odds ratio (OR)=2.70; 95% confidence interval (CI): 1.22-5.99) and were more likely to have underlying HCV (OR=2.91; 95% CI: 1.36-6.23), or alcoholic liver disease (OR=4.20; 95% CI: 1.89-9.35). The only predictors of receipt of therapy were presentation at tumor board conference (OR=2.85; 95% CI: 1.42-5.72) and documented referral to oncology (OR=2.33; 95% CI: 1.10-4.94). CONCLUSIONS: Less than one-third of patients who were diagnosed with HCC received screening before their diagnosis, and of those a large proportion received an alpha-fetoprotein test only. In this study, the use of screening was too suboptimal to be expected to affect outcomes.