Watson, Robert E

AbstractObjectiveTo rigorously compare automated atlas‐based and manual tracing hippocampal segmentation for accuracy, repeatability, and clinical acceptability given a relevant range of imaging abnormalities in clinical epilepsy.MethodsForty‐nine patients with hippocampal asymmetry were identified from our institutional radiology database, including two patients with significant anatomic deformations. Manual hippocampal tracing was performed by experienced technologists on 3T MPRAGE images, measuring hippocampal volume up to the tectal plate, excluding the hippocampal tail. The same images were processed using NeuroQuant and FreeSurfer software. Ten subjects underwent repeated manual hippocampal tracings by two additional technologists blinded to previous results to evaluate consistency. Ten patients with two clinical MRI studies had volume measurements repeated using NeuroQuant and FreeSurfer.ResultsFreeSurfer raw volumes were significantly lower than NeuroQuant (P < 0.001, right and left), and hippocampal asymmetry estimates were lower for both automatic methods than manual tracing (P < 0.0001). Differences remained significant after scaling volumes to age, gender, and scanner matched normative percentiles. Volume reproducibility was fair (0.4–0.59) for manual tracing, and excellent (>0.75) for both automated methods. Asymmetry index reproducibility was excellent (>0.75) for manual tracing and FreeSurfer segmentation and fair (0.4–0.59) for NeuroQuant segmentation. Both automatic segmentation methods failed on the two cases with anatomic deformations. Segmentation errors were visually identified in 25 NeuroQuant and 27 FreeSurfer segmentations, and nine (18%) NeuroQuant and six (12%) FreeSurfer errors were judged clinically significant.InterpretationAutomated hippocampal volumes are more reproducible than hand‐traced hippocampal volumes. However, these methods fail in some cases, and significant segmentation errors can occur.

Objective To describe an expanded phenotypic spectrum and longitudinal outcome in 256 LGI1‐IgG–seropositive and/or CASPR2‐IgG–seropositive patients. Methods Patients were identified through service neural autoantibody evaluation. Ninety‐five had longitudinal follow‐up (7–456 months; median = 35). Results Among 3,910 patients tested, 196 were LGI1‐IgG positive, 51 were CASPR2‐IgG positive, and 9 were dual positive. Cerebrospinal fluid testing was less sensitive than serum testing, detecting only 24 of 38 (63%) LGI1‐IgG–positive and 5 of 6 (83%) CASPR2‐IgG–positive patients. LGI1‐IgG–positive specimens had higher voltage‐gated potassium channel–IgG immunoprecipitation values (0.33nmol/l, range = 0.02–5.14) than CASPR2‐IgG–positive specimens (0.10nmol/l, range = 0.00–0.45, p < 0.001). Of patients presenting with pain or peripheral nervous system (PNS) manifestations, 39% were LGI1‐IgG seropositive (7% had solely neuropathy or pain). Multivariate analysis identified age as the only significant predictor of central nervous system (CNS) versus PNS involvement (>50 years; odds ratio = 15, p < 0.001). Paroxysmal dizziness spells (PDS), a unique LGI1‐IgG accompaniment (14% of patients), frequently delayed the diagnosis. T2‐mesiotemporal hyperintensity was more common in LGI1‐IgG–positive (41%) than in CASPR2‐IgG–positive patients ( p = 0.033). T1‐bright basal ganglia were confined to LGI1‐IgG–positive patients with faciobrachial–dystonic seizures (9 of 39, 31%). Cancer was found in 44% of LGI1‐IgG/CASPR2‐IgG dual seropositive patients (one‐third thymoma). Response to initial immunotherapy was favorable in 97%; mean modified Rankin score was 3 (range = 1–5) at onset and 1.74 (range = 0–6) at last follow‐up, with 9% having severe refractory disability, 20% being asymptomatic, 28% receiving immunotherapy, and 58% receiving antiepileptic medication. Interpretation Older age is a strong predictor of CNS involvement in patients seropositive for CASPR2‐IgG or LGI1‐IgG. Pain, peripheral manifestations, and stereotypic paroxysmal dizziness spells are common with LGI1‐IgG. Response to initial immunotherapy is often favorable, but some patients remain severely disabled, requiring long‐term immunotherapy and/or antiepileptic medications. Ann Neurol 2017;82:79–92