Lynn T. Tanoue


Curr Opin Pulm Med. 2016;22(4):327-335. 

In This Article

Potential Harms of Screening

Balanced against the mortality benefit seen in the NLST, screening trials have also clearly demonstrated that screening incurs potential harms. These include a high rate of false-positive findings, radiation exposure related to further diagnostic imaging studies, 'unnecessary' invasive procedures, overdiagnosis of cancers not destined to be clinically significant, identification of incidental nonpulmonary findings that generate further testing and interventions, the potential for patient psychological distress, and the burden of related costs. Moreover, questions have been raised about the generalizability of the results of RCTs to the overall population. These issues highlight the importance of shared decision-making. An informed discussion of the balance between benefit and risks for an individual patient is a crucial component of screening, and is mandated for the Medicare population in the USA by the Centers for Medicare and Medicaid Services.[6]

The Achilles' heel of screening with LDCT is the high rate of false-positive findings. In NLST, the rate of positive screens, defined as identification of a nodule(s) with diameter at least 4 mm, over the three rounds of screening were 27.3, 27.9, and 16.8%, of which 96.2, 97.6, and 94.8%, respectively, were false-positive findings.[16] The positive predictive value of a positive finding was only 4%. The DLCST and German Lung Cancer Screening Intervention trials used 5 mm as the threshold for an abnormal nodule, but this still resulted in a high rate of positive screens (Table 1).[19,23] Management of positive findings drives much of the remaining sources of potential harm, emphasizing the importance of a rigorous approach to their evaluation.[29]

Radiation exposure related to imaging studies is a concern often raised by patients. The International Commission on Radiologic Protection estimates that 3–6 radiation-induced cancers will occur over a 15–20 year period per 100 000 persons screened.[30] Although these would be far fewer than the number of deaths averted by screening, this highlights the potential impact of even more radiation from diagnostic studies done because of false-positive findings. An LDCT effective radiation dose is approximately 1.5 mSv, compared with 8 mSv for a diagnostic chest CT. We should recognize that diagnostic quality scans are often necessary to make decisions about screen-detected abnormalities.[31] The NLST was remarkable in how few participants underwent invasive procedures for nonmalignant findings. Only 2.7% of the17 053 study participants in the LDCT arm who did not subsequently have lung cancer confirmed, that is, the false-positive population, underwent needle biopsy, bronchoscopy, mediastinoscopy, thoracoscopy, or thoracotomy.[16] Of these procedures, less than 1% were associated with a major complication. Moreover, surgical mortality for patients who did have lung cancer and underwent resection was only 1%, several-fold lower than reported national outcomes.[32]

Overdiagnosis, the detection of an indolent cancer not destined to cause harm, continues to be part of the lung cancer screening discussion. That overdiagnosis exists for lung cancer was supported by the Mayo Lung Project, in which CXR was studied as the screening intervention.[33] Though more lung cancers were diagnosed in the CXR arm, ultimately no benefit in mortality was observed over many years of follow-up. A model evaluating overdiagnosis in the NLST estimated that 18.5% of all lung cancers diagnosed by LDCT were indolent.[34] Although the overall benefit of screening persisted, this points out the need for better tools to identify more or less aggressive disease.

The potential for screening to cause psychological distress is of particular concern because of the high false-positive rate. A recent systematic review evaluating the impact of screening on quality of life, distress, and anxiety demonstrated that screening was associated with transient psychological discomfort in many people, but did not affect health-related quality of life.[35] Although false-positive results were associated with short-term increases in distress, these subsequently returned to levels similar to people with negative screens. Similarly, an analysis of NLST participants demonstrated no difference in quality of life measurements in study participants who had a false-positive screen compared with those whose screens were negative.[36]

One further consequence of screening is the identification of incidental findings other than lung nodules. These include other lung diseases, particularly emphysema and interstitial lung disease, and nonpulmonary findings, most commonly coronary artery calcification, aortic abnormalities, adrenal nodules, and thyroid nodules. The serendipitous finding of a major abnormality, for example, severe coronary artery calcification or aortic aneurysm, may be highly beneficial. However, there is the potential that many findings will not be clinically significant but may trigger subsequent evaluation that is unnecessary or even harmful. The clinical impact of these incidental findings has not been defined, but a standardized approach, similar to the Fleischner Society guidelines for management of incidental pulmonary nodules identified on nonscreening CT scans, should be considered.[37,38]