Eiichiro Iwata1, Hideki Shigematsu2, Yusuke Yamamoto2, Sachiko Kawasaki3, Masato Tanaka2, Akinori Okuda4, Yasuhiko Morimoto2, Keisuke Masuda2, Munehisa Koizumi5, Manabu Akahane6, Yasuhito Tanaka2. 1. Department of Orthopedic Surgery, Nara Medical University, Nara, Japan. Electronic address: iwata@naramed-u.ac.jp. 2. Department of Orthopedic Surgery, Nara Medical University, Nara, Japan. 3. Department of Orthopedic Surgery, Kashiba Asahigaoka Hospital, Nara, Japan. 4. Department of Emergency and Critical Care Medicine, Nara Medical University, Nara, Japan. 5. Department of Spine Surgery, Nara Prefecture General Medical Center, Nara, Japan. 6. Department of Public Health, Health Management and Policy, Nara Medical University, Nara, Japan.
Abstract
BACKGROUND: Preoperative differential diagnosis between spinal meningioma and schwannoma is critical due to the characteristic differences of the surgical treatments. Thus, we aimed to develop an algorithm for the differential diagnosis of these two lesions based on plain MRI findings. METHODS: We retrospectively reviewed plain MR images from patients who had undergone surgical treatment for meningiomas and schwannomas in our hospital between 2002 and 2016. Seven findings characteristic of meningioma or schwannoma were considered: (a) low or equal signal intensity on T2-weighted images, (b) obtuse angle from the dura mater, (c) anterior location in the spinal canal, (d) cystic degeneration, (e) lumbar occurrence, (f) oval or round shape, and (g) dumbbell type. We calculated the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of each MRI finding, following which the algorithm was developed using decision tree analysis. Finally, we examined the usefulness of the algorithm for differential diagnosis between the two lesions. RESULTS: Twenty four patients with meningiomas and 56 with schwannomas were enrolled. The sensitivity, specificity, PPV, NPV, and accuracy of each finding were as follows: (a) [58%, 100%, 100%, 85%, 88%], (b) [67%, 89%, 73%, 86%, 83%], (c) [29%, 88%, 50%, 74%, 70%], (d) [30%, 96%, 94%, 37%, 50%], (e) [43%, 96%, 96%, 42%, 59%], (f) [33%, 88%, 73%, 57%, 60%], and (g) [25%, 96%, 93%, 35%, 46%]. Significant differences were observed with regard to (a), (b), (d), (e), and (g). The algorithm was developed using these five findings, all of which exhibited high specificity and reliability. Accuracy of the algorithm was 91.3%. CONCLUSIONS: Our results indicated that plain MRI findings can be used to differentiate between spinal meningiomas and schwannomas. Furthermore, our novel algorithm exhibited high accuracy, suggesting that this algorithm may aid in the differential diagnosis of these two lesions.
BACKGROUND: Preoperative differential diagnosis between spinal meningioma and schwannoma is critical due to the characteristic differences of the surgical treatments. Thus, we aimed to develop an algorithm for the differential diagnosis of these two lesions based on plain MRI findings. METHODS: We retrospectively reviewed plain MR images from patients who had undergone surgical treatment for meningiomas and schwannomas in our hospital between 2002 and 2016. Seven findings characteristic of meningioma or schwannoma were considered: (a) low or equal signal intensity on T2-weighted images, (b) obtuse angle from the dura mater, (c) anterior location in the spinal canal, (d) cystic degeneration, (e) lumbar occurrence, (f) oval or round shape, and (g) dumbbell type. We calculated the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of each MRI finding, following which the algorithm was developed using decision tree analysis. Finally, we examined the usefulness of the algorithm for differential diagnosis between the two lesions. RESULTS: Twenty four patients with meningiomas and 56 with schwannomas were enrolled. The sensitivity, specificity, PPV, NPV, and accuracy of each finding were as follows: (a) [58%, 100%, 100%, 85%, 88%], (b) [67%, 89%, 73%, 86%, 83%], (c) [29%, 88%, 50%, 74%, 70%], (d) [30%, 96%, 94%, 37%, 50%], (e) [43%, 96%, 96%, 42%, 59%], (f) [33%, 88%, 73%, 57%, 60%], and (g) [25%, 96%, 93%, 35%, 46%]. Significant differences were observed with regard to (a), (b), (d), (e), and (g). The algorithm was developed using these five findings, all of which exhibited high specificity and reliability. Accuracy of the algorithm was 91.3%. CONCLUSIONS: Our results indicated that plain MRI findings can be used to differentiate between spinal meningiomas and schwannomas. Furthermore, our novel algorithm exhibited high accuracy, suggesting that this algorithm may aid in the differential diagnosis of these two lesions.
Authors: Young Il Won; Yunhee Choi; Woon Tak Yuh; Shin Won Kwon; Chi Heon Kim; Seung Heon Yang; Chun Kee Chung Journal: Sci Rep Date: 2022-06-16 Impact factor: 4.996