Background: The National Lung Screening Trial revealed that low dose computed tomography (CT) screening reduced lung cancer mortality by 20%. However, nearly all (96.4%) of the positive screening results were false-positive. A higher false-positive rate (FPR) is expected in Korea, where the prevalence of tuberculosis and parasitic diseases are high.
Material and methods: We retrospectively reviewed the records of 1587 cases (906 males, 57.1%; 495 females, 31.2%) in which chest CT was used for health screening from 2006 to 2011 in one institution. The mean ± standard deviation age of the subjects was 62.7 ± 5.7 years and 495 (31.2%) subjects had a smoking history.
Results: Three hundred and thirty six subjects (21.2%) had non-calcified pulmonary nodules (NCPNs) described as solid nodules (n = 319), masses (n = 15) or pure or mixed ground glass opacities (n = 36). The incidence of NCPNs was 23.8% in smokers and 20.0% in non-smokers (P = 0.08). During a median follow up duration of 37 months (range, 0-67 months), eight subjects were confirmed to have lung cancer. Positive predictive value (PPV) of positive CT screening was 2.4% and FPR was 97.6%. Among 495 subjects who had a smoking history, 118 subjects displayed NCPNs (23.8%) and four patients were diagnosed with lung cancer, with a PPV and FPR of 3.4% and 96.6%, respectively.
Conclusion: CT screening has low PPV and high FPR, even in subjects with a high risk of lung cancer.
Keywords: Computed tomography; early detection of cancer; lung neoplasms; pulmonary nodule.
© 2013 Tianjin Lung Cancer Institute and Wiley Publishing Asia Pty Ltd.