The definitive diagnosis of the type of epilepsy, if it exists, in medication-resistant seizure disorder is based on the efficient combination of clinical information, long-term video-electroencephalography (EEG) and neuroimaging. Diagnoses are reached by a consensus panel that combines these diverse modalities using clinical wisdom and experience. Here we compare two methods of multimodal computer-aided diagnosis, vector concatenation (VC) and conditional dependence (CD), using clinical archive data from 645 patients with medication-resistant seizure disorder, confirmed by video-EEG. CD models the clinical decision process, whereas VC allows for statistical modeling of cross-modality interactions. Due to the nature of clinical data, not all information was available in all patients. To overcome this, we multiply-imputed the missing data. Using a C4.5 decision tree, single modality classifiers achieved 53.1%, 51.5% and 51.1% average accuracy for MRI, clinical information and FDG-PET, respectively, for the discrimination between non-epileptic seizures, temporal lobe epilepsy, other focal epilepsies and generalized-onset epilepsy (vs. chance, p<0.01). Using VC, the average accuracy was significantly lower (39.2%). In contrast, the CD classifier that classified with MRI then clinical information achieved an average accuracy of 58.7% (vs. VC, p<0.01). The decrease in accuracy of VC compared to the MRI classifier illustrates how the addition of more informative features does not improve performance monotonically. The superiority of conditional dependence over vector concatenation suggests that the structure imposed by conditional dependence improved our ability to model the underlying diagnostic trends in the multimodality data.
Suppression of hepatitis B virus (HBV) DNA to undetectable levels is an important goal for HIV/HBV-coinfected patients receiving anti-HBV-active antiretroviral therapy (ART), and current guidelines recommend that this outcome should be reached by one year of treatment. However, the proportion of patients that fail to achieve an undetectable HBV DNA at this time point and its determinants remain unknown in clinical practice. The objective of this study was to determine the incidence and risk factors for incomplete HBV suppression following one year of tenofovir-based ART. We performed a cohort study among tenofovir-treated HIV/HBV-coinfected patients. Patients had HBV viremia, initiated tenofovir-based ART, and had HBV DNA measured at one year of therapy. The primary outcome was incomplete HBV suppression (HBV DNA ≥2.6 log IU/mL) at one year. Logistic regression determined odds ratio (ORs) of incomplete HBV suppression for risk factors of interest. Among 133 patients, 54% (95% CI, 46%–63%) had incomplete HBV suppression at one year. Incomplete suppression was associated with higher baseline HBV DNA (OR, 1.46 per log IU/mL increase; 95% CI, 1.1–1.94) and detectable HIV viremia at one year (OR, 2.52; 95% CI, 1.19–5.32). Among 66 patients with suppressed HIV RNA at one year, 28 (42%) failed to achieve an undetectable HBV DNA. Failure to suppress HBV DNA by one year occurred in a sizeable proportion of tenofovir-treated HIV/HBV-coinfected patients. Higher HBV DNA and detectable HIV viremia were risk factors for incomplete HBV suppression.
hepatitis B virus; HIV/hepatitis B virus coinfection; HIV; tenofovir; lamivudine
Temporal Lobe Epilepsy (TLE) affects brain areas beyond the temporal lobes due to connections of the hippocampi and other temporal lobe structures. Using functional connectivity MRI, we determined the changes of hippocampal networks in TLE to assess for a more complete distribution of abnormality.
Regions of interest (ROIs) were defined in the right and left hippocampi in three groups of participants- left TLE (n=13), right TLE (n=11) and healthy controls (n=16). Brain regions functionally connected to these ROIs were identified by correlating resting-state low-frequency fMRI Blood Oxygenation Level Dependent (BOLD) signal fluctuations. The grouped results were compared using independent sample t-test.
TLE was associated with increased hippocampal connectivity involving several key areas of the limbic network (temporal lobe, insula, thalamus), frontal lobes, angular gyrus, basal ganglia, brainstem and cerebellum along with reduced connectivity involving areas of the sensorimotor cortex (visual, somatosensory, auditory, primary motor) and the default mode network (precuneus). Left TLE had more marked connectivity changes than right TLE.
The observed connectivity changes in TLE indicate dysfunctional networks that underlie widespread brain involvement in TLE. There are identifiable differences in the connectivity of the hippocampi between left and right TLE.
Temporal Lobe Epilepsy (TLE); Hippocampal networks; fMRI; Functional Connectivity; Epilepsy psychopathology; Epileptic Networks
The default mode network (DMN) is composed of cerebral regions involved in conscious, resting state cognition. The hippocampus is an essential component of this network. Here, the DMN in TLE is compared to control subjects to better understand its involvement in TLE. We performed resting state connectivity analysis using regions of interest (ROIs) in the retrosplenium/precuneus (Rsp/PCUN) and the ventro-medial pre-frontal cortex (vmPFC) in 36 subjects (11 with right TLE, 12 with left TLE, 13 controls) to delineate the posterior and anterior DMN regions respectively. We found reduced connectivity of the posterior to the anterior DMN in patients with both right and left TLE. However, the posterior and anterior networks were found to be individually preserved. Lateralization of TLE affects the DMN with left TLE demonstrating more extensive networks. These DMN changes may be relevant to altered cognition and memory in TLE and may be relevant to right vs. left TLE differences in cognitive involvement.
Temporal Lobe Epilepsy (TLE); Default mode network (DMN); fMRI; Functional connectivity; Epilepsy psychopathology; Epileptic networks
Purpose of review
Tremendous advances have occurred in recent years in elucidating basic mechanisms of epilepsy at the level of ion channels and neurotransmitters. Epilepsy, however, is ultimately a disease of functionally and/or structurally aberrant connections between neurons and groups of neurons at the systems level. Recent advances in neuroimaging and electrophysiology now make it possible to investigate structural and functional connectivity of the entire brain, and these techniques are currently being used to investigate diseases that manifest as global disturbances of brain function. Epilepsy is such a disease, and our understanding of the mechanisms underlying the development of epilepsy and the generation of epileptic seizures will undoubtedly benefit from research utilizing these connectomic approaches.
MRI using diffusion tensor imaging provides structural information, whereas functional MRI and electroencephalography provide functional information about connectivity at the whole brain level. Optogenetics, tracers, electrophysiological approaches, and calcium imaging provide connectivity information at the level of local circuits. These approaches are revealing important neuronal network disturbances underlying epileptic abnormalities.
An understanding of the fundamental mechanisms underlying the development of epilepsy and the generation of epileptic seizures will require delineation of the aberrant functional and structural connections of the whole brain. The field of connectomics now provides approaches to accomplish this.
epilepsy; functional connectivity; local circuits; structural connectivity; whole brain
Interictal electroencephalography (EEG) has clinically meaningful limitations in its sensitivity and specificity in the diagnosis of epilepsy because of its dependence on the occurrence of epileptiform discharges. We have developed a computer-aided diagnostic (CAD) tool that operates on the absolute spectral energy of the routine EEG and has both substantially higher sensitivity and negative predictive value than the identification of interictal epileptiform discharges. Our approach used a multilayer perceptron to classify 156 patients admitted for video-EEG monitoring. The patient population was diagnostically diverse with 87 diagnosed with either generalized or focal seizures. The remainder was diagnosed with non-epileptic seizures. The sensitivity was 92% (95% CI: 85–97%) and the negative predictive value was 82% (95% CI: 67%–92%). We discuss how these findings suggest that this CAD can be used to supplement event-based analysis by trained epileptologists.
Epilepsy; machine learning; prediction; non-epileptic seizure; computer aided diagnostics
Interictal FDG-PET (iPET) is a core tool for localizing the epileptogenic focus, potentially before structural MRI, that does not require rare and transient epileptiform discharges or seizures on EEG. The visual interpretation of iPET is challenging and requires years of epilepsy-specific expertise. We have developed an automated computer-aided diagnostic (CAD) tool that has the potential to work both independent of and synergistically with expert analysis. Our tool operates on distributed metabolic changes across the whole brain measured by iPET to both diagnose and lateralize temporal lobe epilepsy (TLE). When diagnosing left TLE (LTLE) or right TLE (RTLE) vs. non-epileptic seizures (NES), our accuracy in reproducing the results of the gold standard long term video-EEG monitoring was 82% [95% confidence interval (CI) 69–90%] or 88% (95% CI 76–94%), respectively. The classifier that both diagnosed and lateralized the disease had overall accuracy of 76% (95% CI 66–84%), where 89% (95% CI 77–96%) of patients correctly identified with epilepsy were correctly lateralized. When identifying LTLE, our CAD tool utilized metabolic changes across the entire brain. By contrast, only temporal regions and the right frontal lobe cortex, were needed to identify RTLE accurately, a finding consistent with clinical observations and indicative of a potential pathophysiological difference between RTLE and LTLE. The goal of CADs is to complement – not replace – expert analysis. In our dataset, the accuracy of manual analysis (MA) of iPET (∼80%) was similar to CAD. The square correlation between our CAD tool and MA, however, was only 30%, indicating that our CAD tool does not recreate MA. The addition of clinical information to our CAD, however, did not substantively change performance. These results suggest that automated analysis might provide clinically valuable information to focus treatment more effectively.
epilepsy; computer-aided diagnosis; mutual information; temporal lobe epilepsy; PET; fluoro-deoxyglucose positron emission tomography; machine learning
Sleep spindles and K-complexes are EEG hallmarks of non-REM sleep. However, the brain regions generating these discharges and the functional connections of their generators to other regions are not fully known. We investigated the neuroanatomical correlates of spindles and K-complexes using simultaneous EEG and fMRI.
EEGs recorded during EEG-fMRI studies of 7 individuals were used for fMRI analysis. Higher-level group analyses were performed, and images were thresholded at Z≥2.3.
fMRI of 106 spindles and 60 K-complexes was analyzed. Spindles corresponded to increased signal in thalami and posterior cingulate, and right precuneus, putamen, paracentral cortex, and temporal lobe. K-complexes corresponded to increased signal in thalami, superior temporal lobes, paracentral gyri, and medial regions of the occipital, parietal and frontal lobes. Neither corresponded to regions of decreased signal.
fMRI of both spindles and K-complexes depicts signal subjacent to the vertex, which likely indicates each discharges’ source. The thalamic signal is consistent with thalamic involvement in sleep homeostasis. The limbic region’s signal is consistent with roles in memory consolidation. Unlike the spindle, the K-complex corresponds to extensive signal in primary sensory cortices.
Identification of these active regions contributes to the understanding of sleep networks and the physiology of awareness and memory during sleep.
electroencephalography (EEG); functional MRI (fMRI); spindles; K-complexes; sleep; non-REM
The medial temporal structures, including the hippocampus and the entorhinal cortex, are critical for the ability to transform daily experience into lasting memories. We tested the hypothesis that deep-brain stimulation of the hippocampus or entorhinal cortex alters memory performance.
We implanted intracranial depth electrodes in seven subjects to identify seizure-onset zones for subsequent epilepsy surgery. The subjects completed a spatial learning task during which they learned destinations within virtual environments. During half the learning trials, focal electrical stimulation was given below the threshold that elicits an afterdischarge (i.e., a neuronal discharge that occurs after termination of the stimulus).
Entorhinal stimulation applied while the subjects learned locations of landmarks enhanced their subsequent memory of these locations: the subjects reached these landmarks more quickly and by shorter routes, as compared with locations learned without stimulation. Entorhinal stimulation also resulted in a resetting of the phase of the theta rhythm, as shown on the hippocampal electroencephalogram. Direct hippocampal stimulation was not effective. In this small series, no adverse events associated with the procedure were observed.
Stimulation of the entorhinal region enhanced memory of spatial information when applied during learning. (Funded by the National Institutes of Health and the Dana Foundation.)
The vertex sharp transient (VST) is an electroencephalographic (EEG) discharge that is an early marker of non-REM sleep. It has been recognized since the beginning of sleep physiology research, but its source and function remain mostly unexplained. We investigated VST generation using functional MRI (fMRI).
Simultaneous EEG and fMRI were recorded from 7 individuals in drowsiness and light sleep. VST occurrences on EEG were modeled with fMRI using an impulse function convolved with a hemodynamic response function to identify cerebral regions correlating to the VSTs. A resulting statistical image was thresholded at Z>2.3.
Two hundred VSTs were identified. Significantly increased signal was present bilaterally in medial central, lateral precentral, posterior superior temporal, and medial occipital cortex. No regions of decreased signal were present.
The regions are consistent with electrophysiologic evidence from animal models and functional imaging of human sleep, but the results are specific to VSTs. The regions principally encompass the primary sensorimotor cortical regions for vision, hearing, and touch.
The results depict a network comprising the presumed VST generator and its associated regions. The associated regions functional similarity for primary sensation suggests a role for VSTs in sensory experience during sleep.
electroencephalography (EEG); functional MRI (fMRI); sleep; vertex sharp transients
The alpha rhythm in the EEG is 8–12 Hz activity present when a subject is awake with eyes closed. In this study, we used simultaneous EEG and fMRI to make maps of regions whose MRI signal changed reliably with modulation in posterior alpha activity. We scanned 11 subjects as they rested with eyes closed. We found that increased alpha power was correlated with decreased MRI signal in multiple regions of occipital, superior temporal, inferior frontal, and cingulate cortex, and with increased signal in the thalamus and insula. These results are consistent with animal experiments and point to the alpha rhythm as an index of cortical inactivity that may be generated in part by the thalamus. These results also may have important implications for interpretation of resting baseline in fMRI studies.
Alpha rhythm; BOLD; EEG; fMRI; Resting state; Simultaneous; SITE
To describe the trial design for the multicenter Early Randomized Surgical Epilepsy Trial (ERSET). Patients with pharmacoresistant epilepsy are generally referred for surgical treatment an average of two decades after onset of seizures, often too late to avoid irreversible disability. ERSET was designed to assess the safety and efficacy of early surgical intervention compared to continued pharmacotherapy.
ERSET is a randomized controlled, parallel group clinical trial with blinded outcome adjudication. Participants are patients with mesial temporal lobe epilepsy (MTLE) over the age of 12 who have had pharmacoresistant seizures for not more than two years and are determined by detailed evaluation to be surgical candidates prior to randomization. The primary outcome measure is seizure freedom in the second year of a two-year follow-up period. Health-related quality of life (HRQOL), neurocognitive function, ancillary outcomes, and adverse events were also measured.
Significant methodological problems addressed by the study design included: recruitment of participants early in the course of epilepsy, establishment of operational definitions for “pharmacoresistant” and “early,” and standardization of diagnostic testing, medical treatment and surgical interventions across multiple centers.
Rigorous trial designs to assess surgical interventions in epilepsy are necessary to provide evidence to guide treatment. This paper is the first of a series; trial results will be reported in subsequent publications.
Class I evidence for surgical effectiveness in refractory temporal lobe epilepsy (TLE) in 2001 led to an American Academy of Neurology practice parameter in 2003 recommending “referral to a surgical epilepsy center on failing appropriate trials of first-line antiepileptic drugs.” We examined whether this led to a change in referral patterns to our epilepsy center.
We compared referral data for patients with TLE at our center for 1995 to 1998 (group 1, n = 83) and 2005 to 2008 (group 2, n = 102) to determine whether these recommendations resulted in a change in referral patterns for surgical evaluation. Patients with brain tumors, previous epilepsy surgery evaluations, or brain surgery (including epilepsy surgery) were excluded.
We did not find a difference between the groups in the duration from the diagnosis of habitual seizures to referral (17.1 ± 10.0 vs 18.6 ± 12.6 years, p = 0.39) or the age at the time of evaluation (34.1 ± 10.3 vs 37.0 ± 11.8 years, p = 0.08). However, there was a difference in the distributions of age at evaluation (p = 0.03) and the duration of pharmacotherapy (p = 0.03) between the groups, with a greater proportion of patients in group 2 with drug-resistant epilepsy both earlier and later in their treatment course. Nonepileptic seizures were referred significantly earlier than TLE in either group or when combined.
Our analysis does not identify a significantly earlier referral for epilepsy surgery evaluation as recommended in the practice parameter, but suggests a hopeful trend in this direction.
= American Academy of Neurology;
= antiepileptic drug;
= Early Randomized Surgical Epilepsy Trial;
= nonepileptic seizures;
= randomized controlled trial;
= temporal lobe epilepsy;
= vagus nerve stimulator.
Acyclovir-resistant herpes simplex virus (HSV) has become increasingly common, particularly among patients with human immunodeficiency virus (HIV). We present a case of acyclovir-resistant HSV treated with intralesional cidofovir.
When discussing AED conversion in the clinic, both the patient and physician perspectives on the goals and risks of this change are important to consider. To identify patient-reported and clinician-perceived concerns, a panel of epilepsy specialists was questioned about the topics discussed with patients and the clinician’s perspective of patient concerns. Findings of a literature review of articles that report patient-expressed concerns regarding their epilepsy and treatment were also reviewed. Results showed that the specialist panel appropriately identified patient-reported concerns of driving ability, medication cost, seizure control, and medication side effects. Additionally, patient-reported concerns of independence, employment issues, social stigma, medication dependence, and undesirable cognitive effects are important to address when considering and initiating AED conversion.
Epilepsy; antiepileptic drugs; conversion; patient preferences.