[Analysis of Growth Curve Type in Pulmonary Nodules with
Different Characteristics]

Background and objective Follow up by computed tomography (CT) and growth evaluation are routine methods for the differential diagnosis of indeterminate pulmonary nodules in clinical practice. Pulmonary nodules with diverse biological behaviors may show different growth patterns and velocities. The aim of this study is to identify the volume growth curve of both benign and malignant pulmonary nodules. This work also intends to determine these nodules' growth patterns and provide evidence for the establishment of a follow-up strategy.

The CT data of 111 pulmonary nodules (54 solid, 57 subsolid) were retrospectively evaluated using 3D volumetric software. All of these nodules have been followed up at least twice. Of these nodules, 35 were confirmed as lung cancers, whereas 5 were confirmed as benign by pathology or histology. Moreover, 71 nodules showed no growth in more than 2 years. Stable nodules were defined as low-risk nodules, as confirmed by reevaluation from experts. On the basis of their densities and diameters, the nodules were classified into four types: benign/low-risk solid nodules, malignant solid nodules (diameter ≤1 cm and >1 cm), benign/low-risk subsolid nodules, and malignant subsolid nodules (diameter ≤1 cm and >1 cm). The follow-up interval time (d) were plotted on the x-axis, and the nodules' volume (mm3) and logarithmic volume were plotted on the y-axis. Two radiologists subjectively determined the type of growth curve. Chi-square test was performed to compare the growth curves of benign/low-risk and malignant nodules.

Of 18 solid cancers, 12 cases (66%) were found with steep ascendant growth curves. Those of 3 cases (16.7%) were flat ascendant, 2 cases (11.1%) slowly ascendant, and 1 (5.56%) case flat. Of 17 subsolid cancers, 8 cases (47.1%) manifested steep ascendant growth curves. Those of 4 cases (23.5%) were slowly ascendant, 3 (17.6%) flat, and 2 (11.8%) descendant-ascendant. Of 36 benign/low-risk solid nodules, 5 cases (13.9%) manifested descendant growth curves, 17 cases (47.2%) flat, 8 cases (21.6%) slowly ascendant, and 6 cases (16.7%) undulate. Of 40 benign/low-risk subsolid nodules, 4 cases (10%) manifested descendant growth curves, 21 cases (52.5%) flat, 9 cases (22.5%) slowly ascendant, and 6 cases (15%) undulate. The distribution of growth curve types significantly differed between benign/low-risk and malignant nodules (χ2=42.4, P<0.01).

The growth curves of lung cancers are heterogeneous. A steep ascendant curve is the main type for lung cancer, with the exception of flat, slowly ascendant, or even descendant curve. A slowly ascendant curve cannot exclude the diagnosis of lung cancer, especially for subsolid nodules.
.