Different features of pulmonary metastases in differentiated thyroid cancer: Natural history and multivariate statistical analysis of prognostic variables

D. Casara, D. Rubello, G. Saladini, G. Masarotto, A. Favero, M. E. Girelli, B. Busnardo

Research output: Contribution to journalArticle

Abstract

We studied 134 patients with differentiated thyroid cancer and pulmonary metastases. All were treated with total or near total thyroidectomy, radioiodine and L-thyroxine. The prognostic value of the following variables in three groups of patients were evaluated by univariate and multivariate analysis: age at diagnosis, sex, histologic type, tumor extension, cervical lymph node metastases, mediastinic metastases, presence of metastases in distant sites other than lungs (multiple distant metastases) and morphological (chest x-rays) and functional (131I uptake) features of lung metastases. Univariate analysis identified patient age (p <0.0001), morphological and functional features of lung metastases (p <0.0001), presence of multiple distant metastases (p <0.0001) and histologic type (p = 0.04) as significant prognostic factors. Multivariate analysis showed only morphological (p = 0.0014) and functional (p <0.0001) features of lung metastases and the presence of multiple distant metastases (p = 0.01) as significant and independent variables. The data show that early (pre- radiological) scintigraphic diagnosis and 131I therapy of lung metastases appear to be the most important elements in obtaining both a significant improvement in survival rate and a prolonged disease-free time interval in these patients.

Original languageEnglish
Pages (from-to)1626-1631
Number of pages6
JournalJournal of Nuclear Medicine
Volume34
Issue number10
Publication statusPublished - 1993

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology

Fingerprint Dive into the research topics of 'Different features of pulmonary metastases in differentiated thyroid cancer: Natural history and multivariate statistical analysis of prognostic variables'. Together they form a unique fingerprint.

  • Cite this