E Ellison1, P Lapuerta, S E Martin. 1. Department of Pathology, University of Southern California, Los Angeles 90033, USA.
Abstract
BACKGROUND: Although characteristic of thyroid papillary carcinoma, psammoma bodies occasionally arise in benign or nonpapillary lesions. Their predictive value in fine-needle aspirates (FNA), without other features suggestive of papillary carcinoma, is uncertain. METHODS: This retrospective study determined the frequency of psammoma bodies and other calcifications in 313 FNAs and 69 resected thyroids; positive predictive value, sensitivity, and specificity were calculated and compared with a triad of characteristic cytologic features (papillary fronds, nuclear ridges/grooves, and intranuclear inclusions). Psammoma bodies from benign and malignant conditions were assessed for differentiating architectural or staining characteristics. RESULTS: Psammoma bodies were found in only 8 of 313 FNAs. Their positive predictive value for papillary carcinoma was 50%, the rest arising from multinodular goiters. By contrast, the positive predictive value of combined cytologic features was 100%. These features were also much more sensitive, present in 80% of papillary carcinomas compared with 14% with psammoma bodies. The psammoma bodies of papillary carcinoma could not be differentiated from those of other lesions either by staining characteristics or by architecture. Other types of calcifications also were present in diverse conditions, and did not discriminate among them. CONCLUSIONS: Isolated psammoma bodies are an unreliable predictor of papillary carcinoma in the absence of cytologic features. In view of the relatively low aggressiveness of papillary carcinoma in most age groups, repeat fine-needle aspiration and clinical correlation are recommended prior to consideration of surgery.
BACKGROUND: Although characteristic of thyroid papillary carcinoma, psammoma bodies occasionally arise in benign or nonpapillary lesions. Their predictive value in fine-needle aspirates (FNA), without other features suggestive of papillary carcinoma, is uncertain. METHODS: This retrospective study determined the frequency of psammoma bodies and other calcifications in 313 FNAs and 69 resected thyroids; positive predictive value, sensitivity, and specificity were calculated and compared with a triad of characteristic cytologic features (papillary fronds, nuclear ridges/grooves, and intranuclear inclusions). Psammoma bodies from benign and malignant conditions were assessed for differentiating architectural or staining characteristics. RESULTS:Psammoma bodies were found in only 8 of 313 FNAs. Their positive predictive value for papillary carcinoma was 50%, the rest arising from multinodular goiters. By contrast, the positive predictive value of combined cytologic features was 100%. These features were also much more sensitive, present in 80% of papillary carcinomas compared with 14% with psammoma bodies. The psammoma bodies of papillary carcinoma could not be differentiated from those of other lesions either by staining characteristics or by architecture. Other types of calcifications also were present in diverse conditions, and did not discriminate among them. CONCLUSIONS: Isolated psammoma bodies are an unreliable predictor of papillary carcinoma in the absence of cytologic features. In view of the relatively low aggressiveness of papillary carcinoma in most age groups, repeat fine-needle aspiration and clinical correlation are recommended prior to consideration of surgery.