• Benedikt H Heidinger, Ursula Nemec, Kevin R Anderson, Daniel B Costa, Sidhu P Gangadharan, Paul A VanderLaan, and Alexander A Bankier.
• Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215. Electronic address: bheiding@bidmc.harvard.edu.
• Acad Radiol. 2017 Nov 1; 24 (11): 1422-1427.

Rationale And ObjectivesThe objective of this study was to quantify the impact of different rounding methods on size measurements of pulmonary nodules and to determine the number of nodules that change management categories as a result of rounding.Materials And MethodsFor this retrospective institutional review board-approved study, we included 503 incidental pulmonary nodules (308 solid and 195 subsolid) from a data repository. Long and short axes were measured. Average diameters were calculated using four different rounding methods (method 1: no rounding; method 2: rounding only the average diameter to the closest millimeter; method 3: rounding only short and long axes; and method 4: rounding short and long axes and the average diameter to the closest millimeter). Nodules were classified for each rounding method according to the 2017 Fleischner Society guideline management categories. Measurements were compared among the four rounding methods using analysis of variance.ResultsWithout rounding, the average nodule diameter was 15.67 ± 5.97 mm. This increased between 0.03  and 0.29 mm using rounding methods 2-4 (range: P < 0.001-0.017). The nodule size was more frequently rounded up (range: 52.1%-77.5%) than rounded down (range: 17.7%-42.5%) using rounding methods 2-4, as compared to no rounding. In the 308 solid nodules, up to 2.9% of the nodules changed management category, whereas none of the 195 subsolid nodules changed category.ConclusionsRounding methods have a small absolute but statically significant effect on nodule size, impacting management category in less than 3% of the nodules. This suggests that, in clinical practice, any rounding method can be used for determining nodule size without substantially biasing individual nodules toward given management categories.Copyright © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

27 keywords...

hide…