Main Article Content

Positive Predictive Value of High-Grade Prostate Imaging and Reporting Data System V2.1 Magnetic Resonance Imaging Findings for Prostate Cancer


Abdudin Heru Mehammed
Alemayehu Bedane Worke
Ashenafi Aberra Buser
Semira Abrar Issa
Takele Menna
Tesfaye Kebede Legesse
Assefa Getachew Kebede

Abstract

BACKGROUND: Prostate cancer is a leading cause of cancer-related mortality among men, second only to lung cancer. Prostate magnetic resonance imaging (MRI) utilizing the Prostate Imaging and Reporting Data System (PI-RADS) v2.1 scoring system effectively stratifies patients by risk and correlates significantly with histopathological outcomes. This study aimed to assess the positive predictive value (PPV) of high-grade PI-RADS v2.1 MRI findings and their correlation with histopathological results from biopsies in patients visiting the interventional radiology unit at St. Paul’s Hospital Millennium Medical College (SPHMMC).


METHODS: A facility-based cross-sectional study was conducted involving patients referred to the SPHMMC interventional radiology unit with high-grade PI-RADS v2.1 MRI findings who underwent TRUS-guided prostate biopsy between January 2023 and April 2024.


RESULTS: Among 105 patients, the PPV was 94.5% for a PI-RADS v2.1 score of 5 and 51.5% for a score of 4. These findings underscore the predictive power of high-grade PI-RADS scores, particularly for score 5 lesions, aiding clinicians in decision-making for further investigations and treatment. Significant correlations were observed between MRI characteristics—such as ill-defined margins, larger size, and extraprostatic extension—and high-grade PI-RADS scores in the peripheral zone (p<0.01).


CONCLUSION: High-grade PI-RADS v2.1 scores exhibit strong positive predictive value for detecting prostate cancer, emphasizing the essential role of multiparametric MRI in diagnosis. Integrating multiparametric MRI findings with clinical and laboratory data can further enhance patient care and outcomes.


Journal Identifiers


eISSN: 2413-7170
print ISSN: 1029-1857