Biomedicines (Apr 2024)

Validating Brain Tumor Reporting and Data System (BT-RADS) as a Diagnostic Tool for Glioma Follow-Up after Surgery

  • Yassir Edrees Almalki,
  • Mohammad Abd Alkhalik Basha,
  • Maha Ibrahim Metwally,
  • Nesma Adel Zeed,
  • Mohamad Gamal Nada,
  • Sharifa Khalid Alduraibi,
  • Ahmed A. Morsy,
  • Rawda Balata,
  • Ahmed Z. Al Attar,
  • Mona M. Amer,
  • Mohamed Abd El-Aziz Mohamed Farag,
  • Sameh Abdelaziz Aly,
  • Ahmed M. Abdelkhalik Basha,
  • Enas Mahmoud Hamed

DOI
https://doi.org/10.3390/biomedicines12040887
Journal volume & issue
Vol. 12, no. 4
p. 887

Abstract

Read online

Gliomas are a type of brain tumor that requires accurate monitoring for progression following surgery. The Brain Tumor Reporting and Data System (BT-RADS) has emerged as a potential tool for improving diagnostic accuracy and reducing the need for repeated operations. This prospective multicenter study aimed to evaluate the diagnostic accuracy and reliability of BT-RADS in predicting tumor progression (TP) in postoperative glioma patients and evaluate its acceptance in clinical practice. The study enrolled patients with a history of partial or complete resection of high-grade glioma. All patients underwent two consecutive follow-up brain MRI examinations. Five neuroradiologists independently evaluated the MRI examinations using the BT-RADS. The diagnostic accuracy of the BT-RADS for predicting TP was calculated using histopathology after reoperation and clinical and imaging follow-up as reference standards. Reliability based on inter-reader agreement (IRA) was assessed using kappa statistics. Reader acceptance was evaluated using a short survey. The final analysis included 73 patients (male, 67.1%; female, 32.9%; mean age, 43.2 ± 12.9 years; age range, 31–67 years); 47.9% showed TP, and 52.1% showed no TP. According to readers, TP was observed in 25–41.7% of BT-3a, 61.5–88.9% of BT-3b, 75–90.9% of BT-3c, and 91.7–100% of BT-RADS-4. Considering >BT-RADS-3a as a cutoff value for TP, the sensitivity, specificity, and accuracy of the BT-RADS were 68.6–85.7%, 84.2–92.1%, and 78.1–86.3%, respectively, according to the reader. The overall IRA was good (κ = 0.75) for the final BT-RADS classification and very good for detecting new lesions (κ = 0.89). The readers completely agreed with the statement “the application of the BT-RADS should be encouraged” (score = 25). The BT-RADS has good diagnostic accuracy and reliability for predicting TP in postoperative glioma patients. However, BT-RADS 3 needs further improvements to increase its diagnostic accuracy.

Keywords