医学
内科学
肝癌
危险系数
入射(几何)
癌症
人口
胃肠病学
肝病
置信区间
物理
环境卫生
光学
作者
Sven H. Loosen,Karel Kostev,Münevver Demir,Mark Luedde,Verena Keitel,Tom Luedde,Christoph Roderburg
标识
DOI:10.1016/j.ejca.2022.03.010
摘要
Non-invasive tests such as the FIB-4 index (calculated on the basis of patients' age, AST/ALT levels and platelet count) have been recommended by different guidelines to rule out advanced fibrosis and stratify the risk of liver-related outcomes in patients with chronic liver diseases. An elevated FIB-4 index ≥1.3 is an independent predictor for liver cancer development among patients with NAFLD. However, the predictive value of an elevated FIB-4 index regarding (liver) cancer development among the general population is unknown.Using the Disease Analyzer database (IQVIA) which compiles diagnoses, laboratory values, and basic medical and demographic data for over 7.5 million patients followed in general practices in Germany, we identified 248,224 patients with available lab values for FIB-4 index calculation between 2005 and 2019. Patients with an average FIB-4 index of <1.3 were matched 1:1 to patients with a FIB-4 index of ≥1.3 by age and sex.Within 10 years of the index date, the cumulative incidence of liver cancer was significantly higher in patients with a FIB-4 index ≥1.3 (0.25% versus 0.04%, p < 0.001). In a multivariate Cox-regression analysis, this association was found to be higher in patients with pre-existing liver disease (hazard ratio, HR: 10.54 versus 3.79). The risk of liver cancer development was most pronounced among men (HR: 9.40, p < 0.001) and patients ≤50 years of age (HR: 16.36, p < 0.001). A persistently high or increasing FIB-4 index over time was a strong predictor for liver cancer. Notably, the cumulative incidence of cancer in general was comparable among patients with a low/high FIB-4 index (15.7% versus 16.6%; p = 0.060).Our data support the use of the FIB-4 index to estimate the risk of liver cancer development even in unselected patients without pre-existing liver disease. By contrast, the index does not appear to have any predictive value for other tumour sites.
科研通智能强力驱动
Strongly Powered by AbleSci AI