Papers

Peer-reviewed
Apr, 2017

Diagnostic Value of Dual-time-point F-18 FDG PET/CT and Chest CT for the Prediction of Thymic Epithelial Neoplasms

ACTA MEDICA OKAYAMA
  • Takayoshi Shinya
  • ,
  • Takashi Tanaka
  • ,
  • Junichi Soh
  • ,
  • Toshi Matsushita
  • ,
  • Shuhei Sato
  • ,
  • Shinichi Toyooka
  • ,
  • Tadashi Yoshino
  • ,
  • Shinichiro Miyoshi
  • ,
  • Susumu Kanazawa

Volume
71
Number
2
First page
105
Last page
112
Language
English
Publishing type
Research paper (scientific journal)
DOI
10.18926/AMO/54978
Publisher
OKAYAMA UNIV MED SCHOOL

We retrospectively assessed the dual-time-point (DTP) F-18 FDG PET/CT findings of thymic epithelial neoplasms (TENs) and investigated the diagnostic capacity of PET/CT compared to that of CT for predicting carcinoma. We calculated the ratio of the standardized uptake value of the tumor and that of the aortic arch (T/M ratio) for both the 90-min early scan and the 2-h delayed scan in 56 TEN patients. We used a multivariate logistic regression (MLR) analysis to estimate the CT features of carcinoma. We compared the diagnostic capacities of PET/CT and chest CT using receiver operating characteristic (ROC) analyses. The ROC curve revealed that the appropriate cut-off T/M ratio value for the highest accuracy was 2.39 with 75.0% accuracy. The area under the curve (AUC) was 0.855. The statistical analyses for DTP scans of 35 TEN patients demonstrated 74.3% accuracy and 0.838 AUC for the early scan versus 82.9% and 0.825 for the delayed scan. The MLR analysis indicated that mediastinal fat infiltration was a predictor of carcinoma. The ROC curve obtained for the model yielded an AUC of 0.853. Delayed scanning could improve the diagnostic capacity for carcinoma. The T/M ratio and mediastinal fat infiltration are predictive of carcinoma with moderate diagnostic accuracy.

Link information
DOI
https://doi.org/10.18926/AMO/54978
PubMed
https://www.ncbi.nlm.nih.gov/pubmed/28420891
Web of Science
https://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=JSTA_CEL&SrcApp=J_Gate_JST&DestLinkType=FullRecord&KeyUT=WOS:000402016700003&DestApp=WOS_CPL
ID information
  • DOI : 10.18926/AMO/54978
  • ISSN : 0386-300X
  • Pubmed ID : 28420891
  • Web of Science ID : WOS:000402016700003

Export
BibTeX RIS