Role of MRI in Differentiating Various Posterior Cranial Fossa Space-Occupying Lesions Using Sensitivity and Specificity: A Prospective Study

Background: Any abnormal space-occupying posterior fossa lesion may directly involve the vital structures like the brain stem, cranial nerves, cerebellum, vertebrobasilar artery, and venous sinuses, which makes the surgical approach and total excision very difficult. Hence for these reasons, precise evaluation of posterior fossa lesion with MRI is a must to visualize the vital structures, which helps in planning and safe surgery. Objective: This study aimed to evaluate the added value of diffusion-weighted imaging and magnetic resonance spectroscopy in the localization, extension, characterization, differentiation of various posterior fossa space-occupying lesions, and correlating with the histopathological result. Materials and methods: This prospective study comprised of 40 patients who were suspected with posterior fossa space-occupying lesions on basis of clinical features or on CT scan. All patients were evaluated using conventional as well as newer MRI techniques using Siemens 1.5 Tesla MRI scanner (Siemens Medical System, Erlangen, Germany). Diffusion-weighted imaging (DWI) was done in all patients and magnetic resonance spectroscopy (MRS) was done in 27 patients. Based on the MRI findings, various posterior fossa lesions were classified as neoplastic or non-neoplastic. The neoplastic lesions were further classified as benign and malignant. The MRI findings were correlated with histopathological findings or follow-up. Statistical analysis: Independent sample t-test was used to compare the mean apparent diffusion coefficient (ADC) values of various posterior fossa space-occupying lesions. Receiver operating characteristic (ROC) curve analysis was done to determine the optimal cut-off mean ADC values and choline/creatinine (Cho/cr) ratios for various benign and malignant posterior fossa tumors. Results: Of 40 patients with posterior fossa lesions, 23 were males and 17 were females with a mean age of 34.67±1.93[SD] years. Metastases were the most common posterior fossa lesions in our study sample and found in seven patients (17.5%) followed by vestibular schwannomas and brainstem gliomas in five patients (12.5%) each, demyelinating lesion in four patients (10%), tubercular abscess in three patients (7.5%), hemangioblastoma, tuberculoma, arachnoid cyst, epidermoid cyst, pilocytic astrocytoma, low-grade glioma in two patients (5%) each, meningioma, medulloblastoma, pyogenic abscess and high-grade glioma in one patient (2.5%) each. The mean ADC value of benign tumors was higher than that of malignant tumors and this difference was found to be significant (p = 0.019). The cut-off ADC value 1.022 x 10-3mm2/s had a sensitivity of 78.6% and specificity of 66.7%. MRS played important role in differentiating neoplastic from non-neoplastic lesions and benign from malignant tumors. The cut-off Cho/cr ratio of 1.25 had a sensitivity of 66.7%, specificity of 85.7% to differentiate benign from malignant tumors. Conclusion: Conventional MRI sequences able to diagnose most of the benign-appearing lesions of posterior fossa, however, adding advanced MRI sequences like diffusion-weighted imaging and MR spectroscopy helps us to differentiate and diagnose various posterior fossa lesions even closer to the actual histopathological diagnosis.


Introduction
The posterior fossa of the brain consists of vital structures like the brainstem, cerebellum, lower cranial nerves, vertebrobasilar arteries, and venous sinuses. The posterior fossa is the spaces above the foramen of the magnum and below the tentorial cerebelli [1]. Computed tomography of the posterior fossa is limited by the beam hardening artifacts and which adversely hamper the quality of images. MRI has played an increasingly important role and is the investigation of choice in patients suspected of harboring lesions in the posterior fossa. T1-weighted images provide excellent anatomical details while T2-weighted images provide pathological information [2].
Conventional MR imaging provides diagnosis with a success rate of 30-90%. However, advanced MRI sequences like diffusion-weighted imaging (DWI), MR perfusion, and magnetic resonance spectroscopy (MRS) can provide immense value in the characterization and diagnosis of various posterior fossa spaceoccupying lesions [3,4].
DWI is helpful in distinguishing between the brain abscesses, necrotic, and cystic neoplasms on MRI. Increased tumor cellularity in neoplasm causes a relative reduction in the apparent diffusion coefficient (ADC) values and that helps in differentiation between the medulloblastoma, ependymoma, and pilocytic astrocytoma [3]. By adding DWI and MR spectroscopy over conventional MRI can negate the need for biopsy in various posterior fossa tumors [3]. DWI is an excellent tool used in differentiating epidermoid from arachnoid cysts even difficult to differentiate with conventional MR sequences [4]. MRS gives information related to cell membrane proliferation, neuronal damage, energy metabolism, and necrotic transformation of brain or tumor tissues. MRS gives information at the molecular level rather than structural and anatomic imaging of tissues and organs by identifying various chemical metabolites [5,6]. MR spectroscopy can differentiate between the non-neoplastic and neoplastic lesions, benign and malignant brain lesions [6,7].
This study aimed to evaluate the added value of diffusion-weighted imaging and MRS in the localization, extension, characterization, differentiation of various posterior fossa space-occupying lesions, and correlating with the histopathological result.

Study design
After approval from the institutional ethics review committee, a hospital-based cross-sectional prospective study was conducted. This study group comprised of 40 patients with CT proven or clinical suspicion of posterior fossa space-occupying lesion presented to the Department of Radiodiagnosis in a tertiary care center of Northeast India from July 2017 to June 2018. Informed consent was obtained from the patient/parents/guardian before undergoing an MRI scan ( Figure 1). Patients with posterior fossa stroke, patients with a history of head trauma, patients with congenital posterior fossa malformation like Dandy-Walker cyst, inadequate or noisy MR spectroscopy, and lesion size less than 10 mm were excluded from the study.

Test method
All patients were subjected to an MRI scan of the brain in a supine position in Siemens Magnetom Avanto 1.5 Tesla MR scanner (Siemens Medical System, Erlangen, Germany). Conventional MR sequences like axial T1WI, T2WI, fluid-attenuated inversion recovery (FLAIR), susceptibility-weighted image (SWI), diffusionweighted image (DWI), sagittal T1W1, and coronal T2WI sequences were obtained using 4-5 mm slice thickness. DWI using echo-planar imaging (EPI) was done in all the patients before contrast administration. Post gadolinium fat-suppressed T1W images were obtained in different planes.

Analysis
Two radiologists, one having experience of neuroradiology for 13 years and another having four years of experience prospectively analyzed the MRI images. The two radiologists were blinded to the clinical history or previous radiological reports of the patients. The size of the posterior fossa space-occupying lesion and ADC values were calculated independently by the two radiologists and mean ADC values were used for the results. The MR spectroscopy parameters especially the choline, N-acetyl aspartate (NAA), lipid, lactate, aminoacid peaks, and choline/creatinine ratio were analyzed independently.

Analysis of Conventional MR Images
We evaluated the following characteristics of a space-occupying lesion in the posterior fossa like lesion sizes, locations, margins, T1WI and T2WI appearances, perilesional edema, and lesion heterogeneity. The posterior fossa lesion size was obtained from the largest dimension of a lesion. The location of a posterior fossa lesion is identified according to the involvement of posterior fossa structures. Margins of spaceoccupying lesion in posterior fossa classified into well-defined and ill-defined. The "well-defined" margin was considered when the margin of space-occupying lesion was differentiated from surrounding structures. An "ill-defined" lesion was considered when the margin of a lesion was ill-defined. Lesion heterogeneity was considered when there were mixed-signal intensities on T1WI or T2W images.

ADC Calculation Analysis
Minimum, maximum, and mean ADC values of a posterior fossa space-occupying lesion were calculated from placing either round or elliptical region of interests (ROIs) over the solid, enhancing, non-necrotic, and/or DWI restricted part of a solid enhancing lesion. In a peripherally enhancing cystic lesion, ROIs were placed in the enhancing wall of the lesions. In the case of cystic lesions and enhancing mural nodules, ROIs were placed in the enhancing mural nodule. ADC values were measured in the operating system console using multiple uniform sizes (area: minimum 10 mm 2 , maximum 50 mm 2 ). We divided the lesion into four quadrants and calculated the ADC values from four quadrants and the final mean ADC value was calculated.
In patients with multiple posterior fossa lesions, the largest lesion was selected for calculation of the mean ADC value.

MR Spectroscopy Data Analysis
Appropriate-sized voxels were chosen to analyze the metabolites with MR spectroscopy. MR spectroscopy spectra were analyzed for various metabolites like choline, creatinine, N-acetyl aspartate, lactate, lipid, alanine, and amino acids. Metabolite ratios like Cho/cr were generated. Mean values of the Cho/cr ratio of different lesions were calculated and compared with a final diagnosis. Mean Cho/cr ratio of neoplastic and non-neoplastic lesions, benign and malignant tumors were then compared and their statistical significance was calculated.

Histopathological Analysis/Final Diagnosis
The final diagnosis was achieved by typical MR imaging findings in n= 15 patients, histopathological diagnosis of posterior fossa space-occupying lesion was established on surgically resected specimens in n=19 patients, and follow-up post-treatment scan in n= 6 patients.

Statistical Analysis
All the statistical analysis was performed using Statistical Package for Social Sciences (SPSS) version 16 (Armonk, NY: IBM Corp.). The mean ADC values of various lesions were calculated by an independent sample t-test. The optimal cut-off mean ADC values and Cho/cr ratio for various posterior fossa tumors were obtained from receiver operating characteristic (ROC) curve analysis.

Results
Of 40 patients with posterior fossa lesions, 23 were males and 17 were females with a mean age of 34.67±1.93[SD] years. Maximum nine patients (22.5%) each were in the age group of 11-20 years and 31-40 years followed by seven patients (17.5%) in the age group of 41-50 years and five patients (12.5%) in the age group of 21-30 years. Only seven patients (17.5%) were above the age of 50 years. Headache and vomiting were the most common presenting symptoms found in 24 patients (60%).
Metastases were the most common posterior fossa lesions in our study sample and found in seven patients (17.5%) followed by vestibular schwannomas and brainstem gliomas in five patients (12.5%) each, demyelinating lesion in four patients (10%), tubercular abscess in three patients (7.5%), hemangioblastoma, tuberculoma, arachnoid cyst, epidermoid cyst, pilocytic astrocytoma, low-grade glioma in two patients (5%) each, meningioma, medulloblastoma, pyogenic abscess and high-grade glioma in one patient (2.5%) each. The different posterior fossa lesions with their mean ADC values are shown in Table 2.

ADC: apparent diffusion coefficient
Posterior fossa metastasis was the most common intra-axial lesion in posterior fossa lesion (Figures 2A-2F) while vestibular schwannoma was the most common extra-axial tumor ( Figures 3A-3F).  FLAIR: fluid-attenuated inversion recovery; DWI: diffusion-weighted imaging; ADC: apparent diffusion coefficient DWI and MR spectroscopy differentiate tubercular (Figures 5A-5F) and pyogenic abscess ( Figures  6A-6F). Pyogenic abscess shows an increased amino acid peak at 0.9 ppm, lipid-lactate peak at 1.3 ppm, and succinate peak at 2.3 ppm ( Figure 6) on MR spectroscopy while tubercular abscesses show increased lipid peak at 0.9 ppm with or without lactate peak at 0.9-1.33 ppm ( Figure 5). Brainstem gliomas showed diffusion restriction with low ADC values with a raised choline, lactate peaks, and Cho/cr ratio on MR spectroscopy ( Figures 7A-7F). Two patients each having cerebellar pilocytic astrocytomas and hemangioblastoma showed minimally raised choline peak with a low Cho/cr ratio (less than 1.2) in MR spectroscopy.  Metastases were the most common intra-axial lesion to affect the posterior fossa in our study followed by brainstem glioma. In our study, metastatic lesions showed reduced NAA level, elevated choline peak, Cho/cr ratio, and prominent lipid-lactate peak which correlated with the study conducted by Ningapa et al. [8] and Meyer et al. [9]. No elevated choline peak was observed in perilesional edema. High-grade glioma shows similar MR spectral findings, however, there were increased choline peak, Cho/NAA, and Cho/cr ratios in perilesional edema and correlated with previous studies by Law et al. [10], Wang et al. [11], Tsougos et al. [12]. The brainstem gliomas and medulloblastoma showed areas of diffusion restriction with low ADC values along with a raise choline peak and Cho/cr ratio in our study sample [13][14][15][16]. Pilocytic astrocytoma and hemangioblastoma did not show diffusion restriction [17].
Vestibular schwannomas diagnosed on basis of its classical location at the opening of the internal auditory canal, extension along the course of the eighth cranial nerve and classic "ice-cream on cone appearance" diagnosed on conventional MR images [18].
Diffusion-weighted imaging with ADC values plays an important role in differentiating arachnoid cyst from an epidermoid cyst in with epidermoid cysts owing to their cellular content exhibiting diffuse restriction with lower to variable ADC values [4,19].
In our study, MR spectroscopy showed prominent amino acid and succinate peaks in pyogenic abscesses, prominent lipid peaks in a tubercular abscess. These MR spectroscopy findings are correlated with previous studies of Karatag and Karatag [20], Gupta et al. [21], Kumar et al. [22], and Meena et al. [23].
Kumar et al. found a good correlation of advanced MRI imaging in the evaluation of posterior fossa tumors with histopathological diagnosis [2]. Darweesh et al. and Moller-Hartmann et al. found a more added value of a combination of MR spectroscopy and ADC values over conventional MR sequences in the differentiation and grading of brain tumors [24,25]. In our study, we found similar observations wherein the use of DWI and