Marco Zoccarato1,2, Silvia Valeggia3, Luigi Zuliani4,5, Matteo Gastaldi6, Sara Mariotto7, Diego Franciotta6, Sergio Ferrari7, Giuseppe Lombardi8, Vittorina Zagonel8, Piera De Gaspari4, Mario Ermani9, Alessio Signori10, Anna Pichiecchio11, Bruno Giometto12, Renzo Manara13. 1. Neurology Unit, AULSS 6 Euganea, Padua, Italy. marcozoccarato@gmail.com. 2. Neuroimmunology Group, Istituto di Ricerca Pediatrica, Padua, Italy. marcozoccarato@gmail.com. 3. Neurology Unit, AULSS 6 Euganea, Padua, Italy. 4. Neuroimmunology Group, Istituto di Ricerca Pediatrica, Padua, Italy. 5. Department of Neurology, Ospedale San Bortolo, AULSS 8 Berica, Vicenza, Italy. 6. Neuroimmunology Laboratory, IRCSS Mondino Foundation, Pavia, Italy. 7. Neurology Unit, Department of Neuroscience, Biomedicine and Movement Sciences, University of Verona, Verona, Italy. 8. Department of Oncology, Oncology 1, Veneto Institute of Oncology IOV-IRCCS, Padua, Italy. 9. Department of Neurosciences (DNS), Statistic and Informatics Unit, School of Medicine, University of Padua, Padua, Italy. 10. Department of Health Sciences, University of Genoa, Genoa, Italy. 11. Department of Neuroradiology, IRCSS Mondino Foundation, Pavia, Italy. 12. Department of Neurology, Ospedale Santa Chiara, Trento, Italy. 13. Neuroradiology, Department of Medicine and Surgery, Sezione di Neuroscienze, University of Salerno, Salerno, Italy.
Abstract
PURPOSE: Radiological hallmark of autoimmune limbic encephalitis (LE) is a hyperintense signal in MRI T2-weighted images of mesial temporal structures. We aimed to identify conventional magnetic resonance imaging (MRI) features that can help distinguish LE from temporal glioma. METHODS: Brain MRIs of 25 patients affected by antibody-positive autoimmune LE, 24 patients affected by temporal glioma (tumor group), and 5 negative controls were retrospectively blindly evaluated in random order. RESULTS: Ten brain MRIs from the LE group were correctly recognized; one additional patient with mesial temporal hyperintensity with anti-AK5 abs LE was wrongly diagnosed as having a tumor. The brain MRIs of the remaining 14 of the 25 patients with LE were judged negative or, in three cases, showed features not typical for LE. In the tumor group, all MRIs showed pathological alterations diagnosed as tumors in 22/24 cases and as LE in two (2/22, 9%). Unilateral lesions were more common in tumors than in neuroradiologically abnormal LE (96% vs. 18%, p < 0.001). T2/FLAIR hyperintensity of the parahippocampal gyrus was associated more with tumor than with LE (71% vs. 18%) (p = 0,009), as T2/FLAIR hyperintensity of extralimbic structures (p = 0.015), edema (p = 0.041), and mass effect (p = 0.015). Maintenance of gray/white matter distinction was strongly associated with LE (91% vs. 17%, p < 0.001). CONCLUSION: Conventional brain MRI is a fundamental tool in the differential diagnosis between LE and glioma. Bilateral involvement and maintenance of gray/white matter distinction at the cortical/subcortical interface are highly suggestive of LE.
PURPOSE: Radiological hallmark of autoimmune limbic encephalitis (LE) is a hyperintense signal in MRI T2-weighted images of mesial temporal structures. We aimed to identify conventional magnetic resonance imaging (MRI) features that can help distinguish LE from temporal glioma. METHODS: Brain MRIs of 25 patients affected by antibody-positive autoimmune LE, 24 patients affected by temporal glioma (tumor group), and 5 negative controls were retrospectively blindly evaluated in random order. RESULTS: Ten brain MRIs from the LE group were correctly recognized; one additional patient with mesial temporal hyperintensity with anti-AK5 abs LE was wrongly diagnosed as having a tumor. The brain MRIs of the remaining 14 of the 25 patients with LE were judged negative or, in three cases, showed features not typical for LE. In the tumor group, all MRIs showed pathological alterations diagnosed as tumors in 22/24 cases and as LE in two (2/22, 9%). Unilateral lesions were more common in tumors than in neuroradiologically abnormal LE (96% vs. 18%, p < 0.001). T2/FLAIR hyperintensity of the parahippocampal gyrus was associated more with tumor than with LE (71% vs. 18%) (p = 0,009), as T2/FLAIR hyperintensity of extralimbic structures (p = 0.015), edema (p = 0.041), and mass effect (p = 0.015). Maintenance of gray/white matter distinction was strongly associated with LE (91% vs. 17%, p < 0.001). CONCLUSION: Conventional brain MRI is a fundamental tool in the differential diagnosis between LE and glioma. Bilateral involvement and maintenance of gray/white matter distinction at the cortical/subcortical interface are highly suggestive of LE.