Diagnostic challenges in patients with temporal lobe seizures and features of autoimmune limbic encephalitis. (30th July 2021)

• Record Type: Journal Article
• Authors: Ismail, Fatme Seval; Spatola, Marianna; Woermann, Friedrich G.; Popkirov, Stoyan; Jungilligens, Johannes; Bien, Christian G.; Wellmer, Jörg; Schlegel, Uwe
• Is Part Of: European journal of neurology. Volume 29:Number 5(2022)
• Page Start: 1303
• Page End: 1310
• Publication Date: 2021-07-30
• Subjects: autoimmune limbic encephalitis -- brain MRI -- diagnostic criteria -- neuronal antibodies -- temporal lobe seizures; Neurology -- Periodicals; Nervous system -- Diseases -- Periodicals; 616.8
• DOI: 10.1111/ene.15026
• Languages: English
• ISSNs: 1351-5101

Abstract

Abstract: Background and purpose: Consensus criteria for autoimmune limbic encephalitis (ALE) allow for a diagnosis even without neuronal antibodies (Abs), but it remains unclear which clinical features should prompt neuronal Ab screening in temporal lobe epilepsy patients. The aim of the study was to investigate whether patients with temporal lobe seizures associated with additional symptoms or signs of limbic involvement may harbor neuronal Abs, and which clinical features should prompt neuronal Ab screening in these patients. Methods: We identified 47 patients from a tertiary epilepsy center with mediotemporal lobe seizures and additional features suggestive of limbic involvement, including either memory deficits, psychiatric symptoms, mediotemporal magnetic resonance imaging (MRI) hyperintensities or inflammatory cerebrospinal fluid (CSF). Neuronal Ab testing was carried out at two independent reference laboratories (Bielefeld‐Bethel, Germany, and Barcelona, Spain). All brain MRI scans were assessed by two reviewers independently. Results: Temporal lobe seizures were accompanied by memory deficits in 35/46 (76%), psychiatric symptoms in 27/42 (64%), and both in 19/42 patients (45%). Limbic T2/fluid‐attenuated inversion recovery signal hyperintensities were found in 26/46 patients (57%; unilateral: n = 22, bilateral: n = 4). Standard CSF studies were abnormal in 2/37 patients (5%). Neuronal Abs were confirmed in serum and/or CSF in 8/47 patients (17%) and were directed against neuronal cell‐surface targets (leucine‐rich glioma inactivated protein 1: n = 1, contactin‐associated protein‐2: n = 1, undetermined target: n = 3) or glutamic acid decarboxylase in its 65‐kD isoform ( n = 3, all with high titers). Compared to Ab‐negative patients, those who harbored neuronal Abs were more likely to have uni‐ or bilateral mediotemporal MRI changes (8/8, 100% vs. 18/38, 47%; p = 0.01, Fisher's exact test). Conclusions: In patients with temporal lobe seizures and additional limbic signs, 17% had neuronal Abs affirming ALE diagnosis. Mediotemporal MRI changes were found in all Ab‐positive cases and had a positive likelihood ratio of 2.11 (95% confidence interval 1.51–2.95). Abstract : In patients with temporal lobe seizures and additional limbic signs, 17% had neuronal antibodies (Abs) affirming diagnosis of autoimmune limbic encephalitis (ALE). Compared to Ab‐negative patients, those who harbored neuronal Abs were more likely to have uni‐ or bilateral mediotemporal magnetic resonance imaging (MRI) changes (8/8, 100% vs. 18/38, 47%; p = 0.01, Fisher's exact test). Accordingly, mediotemporal MRI changes had a positive likelihood ratio of 2.11 (95% confidence interval 1.51–2.95). The predominant neuronal Abs were high‐titer glutamic acid decarboxylase in its 65‐kD isoform (GAD65), leucine‐rich glioma inactivated protein 1 (LGI1) and contactin‐associated protein‐2 (CASPR2). These results suggest that in epilepsy populations with suspected ALE and mediotemporal MRI changes, it is worthwhile to check for neuronal Abs, even if the disease course is chronic.