Incidental findings with cardiac CT evaluation: should we read beyond the heart?

Abstract

Cardiac computed tomographic angiography (CTA) allows for simultaneous evaluation of the lung fields and associated structures. There is a debate as to the benefit of or need for routine overread of the lung fields for incidental findings. The possible improvement in cancer diagnosis with routine overreads is balanced against the major limitations of CT lung screening. Current limitations include (a) a high rate of nodule detection given that >50% of participants may have at least one noncalcified nodule; (b) the increased costs and radiation exposure associated with the resulting follow-up CT scans; (c) the cost and the morbidity of follow-up, including further testing, as well as biopsy or resection of benign noncalcified nodule (at least 25% of such procedures in several trials); (d) a small but difficult to quantify potential risk of cancer associated with multiple follow-up CT scans; and (e) a potential for increased anxiety of both the patient and the physician about nonsignificant pathology. All of these limitations are balanced against a possibility that this could lead to an earlier detection of lung cancer with the consequent improvement in the chances of the patients' survival. Extensive studies of screening CT in older smokers have revealed the prevalence of cancer to be between 0.3 and 1%. However, when applied to an ambulatory population of patients presenting for an evaluation of angina, the prevalence of lung cancer or significant non-cardiac findings may be significantly lower. We have reviewed all the relevant literature and sought to determine the potential benefits and harms of specifically overreading CTA for non-cardiac pathology. The weight of the evidence suggests that it is most prudent to not specifically reconstruct and re-read CTA scans for lung nodules. If a non-cardiac abnormality is visualized by the primary interpreter of the cardiac CT, appropriate referral or follow-up is prudent.