Diagnostic accuracy of diffusion weighted MRI in differentiating benign and malignant meningiomas

• Published in: Pakistan journal of medical sciences Vol. 35; no. 3; pp. 726 - 730
• Main Authors: Sohu, Din Muhammad, Sohail, Saba, Shaikh, Rehana
• Format: Journal Article
• Language: English
• Published: Pakistan Knowledge Bylanes 30-06-2019; AsiaNet Pakistan (Pvt) Ltd; Professional Medical Publications
• Subjects: Accuracy; Age; Brain cancer; Brain tumors; Diagnosis; Driving while intoxicated; Edema; Elderly patients; Females; Histochemistry; Histopathology; Magnetic resonance imaging; Medical diagnosis; Medical imaging; Original; Studies; Surgery; Tumors; Benign meningiomas; Malignant meningiomas; Histopathology; Diagnostic accuracy; DWI
• ISSN: 1682-024X; 1681-715X
• DOI: 10.12669/pjms.35.3.1011

To determine the diagnostic accuracy of DWI in differentiating benign and malignant meningiomas keeping histopathology as gold standard.
This was a descriptive analytical study conducted at Radiology Department, DUHS/Dr. Ruth K. M. Pfau Civil Hospital Karachi, from August 2016 to March 2018.It included152 patients clinically suspected of meningioma on conventional neuroimaging. Imaging features of DWI were compared with histopathology findings. The diagnostic accuracy of DWI was calculated in terms of sensitivity, specificity, accuracy, PPV and NPV using histopathology as gold standard.
There were 59 male and 93 female patients with mean age of 55.38±9.8 years. Mean duration of sign and symptoms was 5.67±2.57 months. Out of 152 patients, 117(77%) and 35(23%) were differentiated into benign and malignant meningiomas respectively by DWI while 135(88.82%) and17(11.18%) patients were diagnosed respectively on histopathology. The sensitivity, specificity, PPV, NPV and accuracy of DWI of 84.4%, 82.3%, 97.4%, 40%, and 84.2% respectively keeping histopathology as gold standard.
DWI features along with calculation of ADC values is a reliable non-invasive technique for differentiating benign and malignant meningiomas. However the low negative predictive value necessitates the use of histopathology.