T1 Weighted Imaging In Central Nervous System Disease: Compared Spin Echo With Inversion Recovery

Objective:To determine the selection sequence of T1-weighted imaging in central nervous system disease by comparing results of spin echo (T1WSE) and inversion recovery (T1WIR).Methods:Gadopentetate dimeglumine (Gd-DTPA) was diluted with deionized water in glass tubes, the concentration of solution was ranged from 0.025 to 5 mmol/L. The solutions were imaged with T1WIR and T1WSE imaging, and signal intensities were measured with a region of interest (ROI) analysis of solutions, white matte (WM) and gray matter (GW) of patients. The solution-to-WM contrast ratio (CR) and solution-to-GW CR were calculated. A phantom in which Gd-DTPA was 1:1 diluted from 0.025 to 0.000024 mmol/L was developed to determine the lowest concentration at which the effects of Gd-DTPA were evident on T1WIR and T1WSE images in vitro. 196 patients referred for imaging of the brain were divided into four groups: A group contained 92 patients with primary brain tumors, B group contained 31 patients with brain metastases, C group contained 24 patients with meningitis, and D group contains 49 patients with various diseases which not contained in the others groups. All patients were underwent axial T1WSE and T1WIR imaging before Gd-DTPA injection. A to C groups and 24 patients in D group were performed for contrast enhancement with single dose Gd-DTPA, and T1WIR and T1WSE were used alternately as the first post-enhanced sequence. Signal intensities were measured with a region of interest (ROI) analysis of the tumor, WM, GM, cerebrospinal fluid (CSF) and tumor on two sequences images before and after contrast medium injection. The SD of noise was measured. The data were used to calculate WM-to-GW CR and contrast-to-noise ratio (CNR), contrast-to-enhancement ratio (CER) and signal-to-noise ratio (SNR) of tumors, CR and CNR of tumor-to-WM, tumor-to-GW and tumor-to-CSF. The tumor volumes were measured on two sequences post-contrast images. T1WSE and T1WIR images were compared for number of brain metastases, conspicuity and boundary of lesions, the definition of tumor from edema, and dural tail sign of meningioma.Results:The phantom test revealed that T1WIR showed a threshold contrast signal at 0.0016 mmol/L as compared with T1WSE at 0.00078 mmol/L. The signal of solution was at peak on T1WIR image with 2mmol/L, but on T1WSE with 3mmol/L. The CER of solution was higher on T1WIR images than on T1WSE images when concentration was ranged from 0.1 to 3 mmol/L. However, when concentration was out of that, the converse was correct. The solution-to-WM and solution-to-GW CR was higher on T1WSE images compared to T1WIR images when concentration ranged from 0.025 to 5 mmol/L. In all of patients, the CR and CNR of WM-to-GM were higher on T1WIR images than on T1WSE images.In A group, the SNR of tumors were 75.4±14.3 and 192.8±45.9 on pre- and post-enhanced T1WIR images,which were higher than 48.7±8.9 and 133.4±41.5 on pre- and post-enhanced T1WSE images (P<0.01), respectively. However, the CER of tumors was lower on T1WIR images than on T1WSE images (164.2±63.4 vs 174.7±68.5). Prior to enhancement, all of tumor-to-background CRs on T1WIR images were higher than on T1WSE images. Following contrast enhancement, CRs of tumor-to-WM and tumor-to-GM were higher on T1WSE images, while tumor-to-CSF CR was higher on T1WIR images. All of tumor-to-background CNRs were higher on T1WIR images before and after contrast medium injection. Pre-enhanced T1WIR images showed superior conspicuity in 62 primary brain tumors (67.4%), equal conspicuity in 26 cases (28.3%), and inferior conspicuity in 4 cases (4.3%). Post-enhanced T1WIR images showed equal conspicuity in 86 cases (93.5%), and inferior conspicuity in 6 cases (6.5%). In delineation of tumor boundaries, pre-enhanced T1WIR images were superior in 43 cases (46.7%), equal in 47 cases (51.1%) and inferior in 2 cases (2.2%); post-enhanced T1WIR images were superior in 1 case (1.1%), equal in 75 cases (81.5%) and inferior in 16 cases (17.4%). In 20 cases with edema, the definition of tumor from edema, pre-enhanced T1WIR images were superior in 7 cases (35%) and equal in 13 cases (65%), post-enhanced T1WIR images were superior in 5 cases (25%) and equal in 15 cases (75%). In manifestation of dural tail sign of meningioma, post-enhanced T1WIR images were equal in 9 cases (60%), and inferior in 6 cases (40%). The volume of primary brain tumor was 26.3 cm3±23.6 and 26.3 cm3±23.4 on post-enhanced T1WIR and T1WSE images, respectively. There had no difference between two sequences.In B group, pre-enhanced T1WIR and T1WSE images discovered 94 and 162 lesions, respectively; post-enhanced T1WIR and T1WSE images discovered 236 and 347 lesions, respectively. The SNR of tumors was higher on T1WIR than on T1WSE images before and after contrast medium injection, while the CER of lesions had no difference between two sequences. Before enhancement, all of CRs and CNRs of tumor-to-background were higher on T1WIR images. After enhancement, CRs of tumor-to-WM and tumor-to-GM, and CNR of tumor-to-WM were higher on T1WSE images, while CR of tumor-to-CSF, and CNRs of tumor-to-GM and tumor-to-CSF were higher on T1WIR images. The calculated mean tumor volume had no different between post-enhanced T1WIR and T1WSE images (4.9 cm±6.7 vs 4.8 cm3±6.7).In C group, both post-enhanced T1WIR and T1WSE images were positive in all of tuberculosis, purulent and cysticercal meningitis cases. In all of 17 cases with lymphocytic meningitis, post-enhanced T1WIR images were positive in 6 cases (35.3%), while post-enhanced T1WSE images were positive in 10 cases (58.8%). In 13 subjects who were positive on both sequences post-enhanced images, post-enhanced T1WIR images were equal in 4 cases (30.8%) and inferior in 9 cases (69.2%).In D group, pre-enhanced T1WIR were superior in 17 cases (34.7%), equal in 23 cases (46.9%), and inferior in 4 cases (8.2%); in 5 cases(10.2%), each sequence had partial superiority. According to lesions signal intensity on T1WSE images, both sequences showed equal conspicuity in obvious hypo-intensity signal lesions; T1WIR images showed superior or equal conspicuity in 29 hypo-intensity signal lesions (96.7%); T1WSE images showed superior or equal conspicuity in hyper-intensity lesions; in 8 mingle signal lesions, T1WIR images showed superior conspicuity in 7 hypo-intensity signal lesions and equal conspicuity in 1 hypo-intensity signal lesion, while T1WSE images showed superior conspicuity in 6 hyper-intensity signal hemorrhage lesions and equal conspicuity in 2 hyper-intensity signal hemorrhage lesions. T1WSE images showed superior conspicuity in scolex of 1 brain cysticercosis case. In 20 cases with enhancement on post-enhanced T1WIR and/or T1WSE images, post-enhanced T1WIR images showed equal conspicuity in 15 cases (75%); post-enhanced T1WIR images showed inferior conspicuity in"ivy sign"of 3 Moyamoya disease and 1 brain cysticercosis case; and 1 venous malformation was clearly displayed on post-enhanced T1WSE images, but not on post-enhanced T1WIR images.Conclusion: T1WSE images are more sensitive to lowere concentration Gd-DTPA than T1WSE images do, and have higher solution-to-WM and solution-to-GW CR. In the detection of primary brain tumors, pre-enhanced T1WIR imaging is superior or equal to pre-enhanced T1WSE imaging, while post-enhanced T1WSE imaging is superior or equal to post-enhanced T1WIR imaging. In the detection of brain metastases, T1WIR imaging is inferior to T1WSE imaging before and after contrast medium injection. In revealing meningitis, especially lymphocytic meningitis, post-enhanced T1WIR imaging is inferior to post-enhanced T1WSE imaging. In the detection of the others lesions, T1WIR imaging is superior in revealing hypo-intensity signal lesions, while T1WSE imaging is superior in revealing hyper-intensity signal hemorrhage and abnormal small vascular enhancement.