RSNA 2010 

Abstract Archives of the RSNA, 2010


SSK12-08

Thyroid Nodule Microcalcifications: It’s Harder Than You Think

Scientific Formal (Paper) Presentations

Presented on December 1, 2010
Presented as part of SSK12: Neuroradiology/Head and Neck (Thyroid)

Participants

Abhinav Vij MBBS, MPH, Presenter: Nothing to Disclose
Raul Nirmal Uppot MD, Abstract Co-Author: Nothing to Disclose
Danai Chasaki, Abstract Co-Author: Nothing to Disclose
Elkan F. Halpern PhD, Abstract Co-Author: Research Consultant, Hologic, Inc
Peter Raff Mueller MD, Abstract Co-Author: Consultant, Cook Group Incorporated
Anthony Edward Samir MD, Abstract Co-Author: Nothing to Disclose

PURPOSE

Microcalcifications are a specific signs of thyroid nodule malignancy. However, diagnostic criteria for microcalcification do not exist. This study aims to determine (1) interobserver variability for microcalcification detection, (2) sensitivity and specificity of microcalcification detection for papillary thyroid cancer (PTC), and (3) which microcalcification morphologic characteristics are predictive of PTC.

METHOD AND MATERIALS

Institutional databases were searched to identify subjects between 2002 and 2010 with thyroid nodule microcalcifications, and a cytologic or pathologic diagnosis within one year. Two radiologists reviewed the sonograms independently – a reviewer who reads > 1,500 thyroid sonograms and performs ~400 thyroid biopsies/yr and another reviewer who reads < 500 thyroid sonograms and performs fewer than 50 thyroid biopsies/yr. The reviewers gave binary “absent/present” response to: (1) microcalcifications, (2) macrocalcifications, (3) varying sizes, (4) punctuate, (5) linear, (6) brighter than echoes at the same level elsewhere, (7) anechoic space anterior to the bright echo, and whether the echogenic foci were (8) parallel or variable in orientation, (9) peripheral or central, (10) more or less than 5 and (11) subjectively clustered or distributed. Logistic regression analysis was performed to test the predictive yield of each variable for PTC.

RESULTS

38 subjects had 40 nodules. 23/40 had confirmed PTC on pathology. The sensitivity of the experienced reader for PTC on the basis of detecting microcalcifications was 62% and the specificity was 79%. The sensitivity of the less experienced reader was 33% and specificity was 79%. The kappa coefficient of agreement between the 2 readers was .19. Only the experienced reader was able to diagnose PTC by detecting microcalcifications (p=.012). The only morphologic characteristic that predicted PTC was variable orientation of the echogenic foci (p=.046).

CONCLUSION

Considerable interobserver variation exists in microcalcification detection, and observer experience is needed to accurately diagnose PTC. The only morphologic characteristic predictive of PTC was varying orientation of echogenic foci.

CLINICAL RELEVANCE/APPLICATION

The sonographic diagnosis of microcalcifications remains difficult. Variable orientation of echogenic foci is the most important sign.

Cite This Abstract

Vij, A, Uppot, R, Chasaki, D, Halpern, E, Mueller, P, Samir, A, Thyroid Nodule Microcalcifications: It’s Harder Than You Think.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9013894.html