Screening studies on conventional chest X‑rays and on sputum cytology did not show a reduction in lung cancer mortality. However, screening by low-dose computed tomography (LDCT) is more promising because it allows tumor detection in early stages at fairly low radiation levels. No reduction of lung cancer mortality was found in two small, randomized clinical studies on LDCT screening in Europe. However, in the by far largest LDCT trial, the National Lung Screening Trial (NLST) in the USA, a relative reduction of lung cancer mortality by 20.0% (95% confidence interval: 6.8-26.7%), and a relative reduction of total mortality by 6.7% (95% CI: 1.2-13.6%) was reported. According to the NLST, an important disadvantage of LDCT is the low positive predictive value of abnormal screening results: lung cancer was confirmed in only 4 of 100 abnormal screening results.In this paper, benefits and disadvantages of LDCT screening and related open questions are systematically discussed. A possible reduction of lung cancer specific and total mortality must be weighed against false positive results, overdiagnoses, and radiation exposure. In NLST, the proportion of overdiagnoses is estimated to be 11.0 to 18.5%, depending on the strategy of analysis; radiation exposure is about 1.5 mSv per scan, and thus much lower than radiation exposure in chest X‑ray, which is about 8 mSv per scan. Open questions refer to who should be offered the screening, how long the time intervals between screening rounds should be, and which algorithms should be used to clarify screen-detected nodules.
Keywords: Early detection of cancer; False-positive results; Low-dose computed tomography; Overdiagnoses; Radiation exposure.