Administracion

Imaging of Incidental Findings on Thoracic Computed Tom o g r a p h y
Jeffrey B. Alpert, MD*, David P. Naidich, MD
KEYWORDS
 Incidental  Nodule  Thymoma

IMAGING OF INCIDENTAL FINDINGS ON THORACIC COMPUTED TOMOGRAPHY
Computed tomography (CT) of the chest remains one of the most commonly used tools in the assessment of thoracic disease. With continued technologic advancement ofmultidetector CT (MDCT), ultrathin 1 mm or less volumetric sections of the thorax can be obtained during a single breath-hold. As a result, with higher spatial resolution and greater overall sensitivity, MDCT of the chest has produced a greater number of findings, many of which are unsuspected and of uncertain clinical significance. Among nearly 200 CT angiograms performed in an emergency setting forassessment of pulmonary embolism, Hall and colleagues1 found that patients were more than twice as likely to demonstrate a new incidental finding (24%) that required follow-up than pulmonary embolism (9%). An incidental finding may be considered as any finding that is unsuspected or unrelated to the clinical indication for imaging. When discovered, incidental findings must be categorized asclinically significant or clinically insignificant. If significant, one must determine if immediate action should be taken (such as a newly discovered malignancy), if the finding needs to be recognized and reassessed in time (such as a nonspecific lung nodule), or if the finding needs to be recognized without further work-up (such as variant vascular anatomy).

Jacobs and colleagues2 examined 11screening studies for lung cancer and coronary artery disease and found a wide range of reported incidental thoracic findings. Lung cancer screening studies reported an average of 14.2% of patients with significant incidental findings, compared with 7.7% of patients undergoing coronary artery screening, a difference attributed to the limited field of view used on cardiac CT. There was considerablevariation among all screening studies with unexpected findings that required follow-up, ranging from 3% to 41.5%. Furthermore, recommendations for further evaluation varied widely. These authors found that although there was consistency regarding the definitions used to classify incidental findings, there was a lack of uniformity regarding both the clinical significance and recommendations ascribed tothe findings. Although there are well-established guidelines for follow-up of some incidental findings such as small solid lung nodules, there is no clear follow-up algorithm for many other unexpected findings.3 As a result, with an increasing number of incidental findings, there are now twice as many additional imaging studies recommended when compared to over 10 years ago.4 This increase isoften associated with additional exposure to ionizing radiation and likely a greater risk of radiation-induced malignancy. Ultimately, incidental findings can also become a significant source of medical cost, patient anxiety, and confusion. This article discusses and illustrates

No grant funding or other support was provided for this work. Thoracic Imaging Section, Department of Radiology, NYULangone Medical Center, 560 First Avenue, IRM 236, New York, NY 10016, USA * Corresponding author. E-mail address: Jeffrey.Alpert@downstate.edu Radiol Clin N Am 49 (2011) 267–289 doi:10.1016/j.rcl.2010.10.005 0033-8389/11/$ – see front matter Ó 2011 Elsevier Inc. All rights reserved.

radiologic.theclinics.com

268

Alpert & Naidich
the spectrum of the most commonly encountered incidentalfindings on thoracic CT studies, as well as attempts to differentiate those for which additional imaging or clinical correlation is required from those for which additional evaluation is unwarranted. a reliable indicator of hamartoma, provided there is no history of fat-containing malignancy such as liposarcoma or renal cell carcinoma.9 Hamartomas are also composed of fibrous tissue, epithelial...