| ).
Marginal spiculation has been known for many years to be associated with malignancy, and more recent studies have confirmed spiculation as a risk factor for cancer (7,61). Unfortunately, spiculation has generally been classified in a binary manner as present or absent, and the threshold for determining the presence of spiculation has not been defined. Nonetheless, it has been consistently identified as a risk factor for malignancy, with an odds ratio in the range of 2.2–2.5 in screen-detected nodules (7). Nodule LocationLung cancers occur more frequently in the upper lobes, with a predilection for the right lung (62,63). In the PanCan trial, upper lobe nodule location was confirmed as a risk factor, with an odds ratio of approximately 2.0 (7). Adenocarcinomas and metastases tend to be located in the periphery, while squamous cancers are more often found near the hila (62). Small solid nodules in a perifissural or subpleural location often represent intrapulmonary lymph nodes (discussed later in this article). Nodule MultiplicityAn analysis of patients with multiple nodules in the NELSON trial showed increased risk of primary cancer as the total nodule count increased from 1 to 4 but decreased risk in patients with 5 or more nodules, most of which likely resulted from prior granulomatous infection (29). In the PanCan trial, multiplicity of nodules was associated with a reduced risk of cancer when compared with risk associated with one nodule (7). Nodule Growth RateCancers have a wide range of growth rates that depend on morphology and histologic findings. Recommended follow-up intervals are intended to minimize the number of examinations and the chance of a growing cancer advancing in stage during the period of CT follow-up prior to diagnosis. Thus, we must consider the potential growth rate of a detected nodule and our ability to detect small changes in size when we make a recommendation for follow-up. Although linear measurement with electronic calipers remains the current standard of practice, experience with semiautomated nodule volumetry suggests that this approach has superior sensitivity in the detection of nodule growth (22,64,65 |