Correlation between Magnetic Resonance Imaging (MRI) Findings with Histological Grading of Patients with Supratentorial Diffusely Infiltrating Astrocytomas

Abstract

Data regarding the impact of MRI for diagnosis and staging of brain tumors in our population is sparse. Thus, the purpose of this study is to determine the efficacy of MRI in preoperative diagnosis of primary intra-axial brain tumors and to determine its diagnostic accuracy in grading of gliomas in our population, correlated with histopathological (hematoxylin & eosin stain) findings taking as gold standard. There has been extensive works in developed countries regarding the diagnostic effectiveness of MRI. But only a few works have been done in our country. To be very particular, little work has been done regarding the performance tests of MRI regarding histological correlation with gliomas in Neurosurgery Department, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh. So, I have tried to correlate MRI findings with the histopathological (hematoxylin & eosin stain) findings of supratentorial diffusely infiltrating astrocytomas. To evaluate the relationship of Magnetic Resonance Imaging (MRI) findings with histological grading of patients with gliomas, to assess MRI findings of gliomas and to assess histological grading of gliomas. The study was carried out in the Department of Neurosurgery, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh, from March 2015 to August 2016. This study was descriptive type of observational study (Cross sectional study). Study Population were include all the patients with supratentorial gliomas admitted in the department of neurosurgery, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh, during the above mentioned time. All patients were admitted through out patient & emergency department of Neurosurgery, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh, were included in the study after primary screening with inclusion and exclusion criteria. History of the patients were taken with demographic data. General examinations were carried out after admission. Neurological examinations were carried out thoroughly. Grading of gliomaswere done from MRI findings. Histopathologies were done of resected tumour tissues. After ethical clearance for the study from the Department of Neurosurgery and IRB, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh, written informed consents were taken from the patients and/or the legal guardians/responsible family members after completely explaining the procedure and the purpose of the study in easy local language. Patients data were collected in data collection sheets. Every patient was enjoy every right to participate or refuse participation and had the right to withdraw from the study at any time without compromising their medical care. The privacy of the patients were strictly maintained and the patients informations were notdisclosed to any source. The study data were onlyused for the purpose of this scientific study. Data were processed and analyzed using computer software SPSS (Statistical Package for Social Sciences) version 22.Descriptive data presented on categorical scale were expressed as frequencies and corresponding percentages, while the data presented on continuous scale were measured as mean and standard deviation (SD). Correlation between categorical variables were evaluated using Spearman’s rank correlation test. For analytical tests, level of significance was set at 0.05 and p-value <0.05 was considered significant. The summarized informations were then presented in form of tables and figures. For the validity test, sensitivity, specificity, accuracy, positive predictive value and negative predictive value of MRI in detecting the grade of intracranial gliomas were calculated after confirmation of the diagnosis by histopathology. Results: The age range of patients were 18 to 70 years. The peak age incidence were >50 years age group. 75.8% patients were male and 24.2% patients were female. The male to female ratio was 3.12:1. Regarding midline crossing in MRI of brain , 24 (72.7%) patients showed no crossing of midline, 2 (6.1%) patients showed equivocal and 7 (21.2%) patients showed crossed midline. In 4 (12.1%) patients there were mild edema, 12 (36.4%) patients showed moderate edema and 17 (51.5%) patients showed marked edema. 10 (30.3%) patients had well circumscribed lesion, 15 (45.5%) patients had poorly circumscribed lesion and 8 (24.2%) patients had highly infiltrating lesion. 7 (21.2%) patients showed mild mass effect, 18 (54.5%) patients had moderate mass effect and 8 (24.2%) patients had severe mass effect. Necrosis/cystic change of tumor were present in 19 (57.6%) cases. 6 (18.2%) patients showed mild heterogeneity, 22 (66.7%) patients showed moderate heterogeneity and 5 (15.2%) patients showed severe heterogeneity. Definitive haemorrhage showed 5 (15.2%) patients. WHO grading of gliomas by MRI. 17 (51.5%) patients of the gliomas were classified as WHO Grade IV, followed by 9 (27.3%) patients were Grade II and 7 (21.2%) patients were Grade III. WHO grading of gliomas by histopathology. 16 (48.5%) patients of the gliomas were classified as WHO Grade IV, followed by 9 (27.3%) patients were Grade II and 8 (24.2%) patients were Grade III. Out of 33 patients, 09 patients were diagnosed as grade II (Low grade gliomas) by MRI, among them 08 cases were proved as grade II (Low grade glioma) and 01 case was grade III (Anaplastic Astrocytoma) by histopathology. 07 patients were diagnosed as grade III (Anaplastic Astrocytoma) by MRI, among them 05 cases were proved as grade III (Anaplastic Astrocytoma) and 01 case was grade II (Low grade glioma) and 01 case was grade IV (Glioblastoma Multiforme) by histopathology. 17 patients were diagnosed as grade IV (Glioblastoma Multiforme) by MRI, among them 15 cases were proved as grade IV (Glioblastoma Multiforme) and 02 cases were (Anaplastic Astrocytoma) by histopathology. Among 33 patients, 09 patients were diagnosed as low grade gliomas by MRI, among them true positive were in 08 cases and 01 cases were diagnosed as low grade gliomas by MRI but was diagnosed as high grade gliomas by histopathology. This 01 case was false positive. Out of 33 cases of gliomas 24 were diagnosed as high grade gliomas by MRI, among them 23 cases were true negative and 01 case was diagnosed as high grade gliomas by MRI but was proved as low grade gliomas by histopathology. This 01 case was false negative. MRI is accurate in preoperative diagnosis and assessing the characteristics of intracranial gliomas. It is very accurate in assessing the grade of gliomas. Tumor necrosis, irregular margins, mass effect of tumor and peritumoral edema are most important indicators of tumor grade. The present study revealed that there is significant correlation between the MR imaging features and histopathological grading of intracranial gliomas.

Bang. J Neurosurgery 2020; 10(1): 67-74