Effect of Training on Visual Identification of High Frequency Oscillations-A Delphi-Style Intervention

Objective

We examined the effect of a simple Delphi-method feedback on visual identification of high frequency oscillations (HFOs) in the ripple (80–250 Hz) band, and assessed the impact of this training intervention on the interrater reliability and generalizability of HFO evaluations.

Methods

We employed a morphology detector to identify potential HFOs at two thresholds and presented them to visual reviewers to assess the probability of each epoch containing an HFO. We recruited 19 board-certified epileptologists with various levels of experience to complete a series of HFO evaluations during three sessions. A Delphi-style intervention was used to provide feedback on the performance of each reviewer relative to their peers. A delayed-intervention paradigm was used, in which reviewers received feedback either before or after the second session. ANOVAs were used to assess the effect of the intervention on the reviewers' evaluations. Generalizability theory was used to assess the interrater reliability before and after the intervention.

Results

The intervention, regardless of when it occurred, resulted in a significant reduction in the variability between reviewers in both groups (p_GroupDI = 0.037, p_GroupEI = 0.003). Prior to the delayed-intervention, the group receiving the early intervention showed a significant reduction in variability (p_GroupEI = 0.041), but the delayed-intervention group did not (p_GroupDI = 0.414). Following the intervention, the projected number of reviewers required to achieve strong generalizability decreased from 35 to 16.

Significance

This study shows a robust effect of a Delphi-style intervention on the interrater variability, reliability, and generalizability of HFO evaluations. The observed decreases in HFO marking discrepancies across 14 of the 15 reviewers are encouraging: they are necessarily associated with an increase in interrater reliability, and therefore with a corresponding decrease in the number of reviewers required to achieve strong generalizability. Indeed, the reliability of all reviewers following the intervention was similar to that of experienced reviewers prior to intervention. Therefore, a Delphi-style intervention could be implemented either to sufficiently train any reviewer, or to further refine the interrater reliability of experienced reviewers. In either case, a Delphi-style intervention would help facilitate the standardization of HFO evaluations and its implementation in clinical care.

Regional hippocampal diffusion abnormalities associated with subfield-specific pathology in temporal lobe epilepsy

OBJECTIVE

Hippocampal sclerosis (HS) is the most common pathology and best predictor of surgical outcome for medically refractory patients with temporal lobe epilepsy (TLE). Current clinical MRI methods can detect HS, but subfield pathology is poorly characterized, limiting accurate prediction of seizure-free outcomes after surgery. Diffusion tensor imaging (DTI) can probe regional microstructural changes associated with focal hippocampal pathology, but is typically limited by low-resolution whole-brain acquisitions.

METHODS

High-resolution (1 × 1 × 1 mm3) DTI, T1, and quantitative T2 of the hippocampus was acquired in 18 preoperative TLE patients and 19 healthy controls. Diffusion images were qualitatively assessed for loss of internal architecture, and whole-hippocampus diffusion, volume, and quantitative T2 were compared across groups. Regional hippocampal diffusion abnormalities were examined in all subjects and compared to histology in four subjects who underwent anterior temporal lobectomy.

RESULTS

High-resolution mean diffusion-weighted images enabled visualization of internal hippocampal architecture, used to visually identify HS with 86% specificity and 93% sensitivity. Mean diffusivity (MD) elevations were regionally heterogenous within the hippocampus and varied across TLE patients. The spatial location of diffusion abnormalities corresponded with the location of focal subfield neuron loss, gliosis, and reduced myelin staining abnormalities identified with postsurgical histology in four subjects who underwent anterior temporal lobectomy. Whole-hippocampus MD and T2 relaxation times were higher, and fractional anisotropy (FA) and volumes were lower in TLE patients relative to controls. Left hippocampus MD correlated with verbal memory in the TLE group.

SIGNIFICANCE

Visualization of internal architecture and focal diffusion abnormalities on high-resolution diffusion imaging suggests potential clinical utility of diffusion imaging in TLE and may have significant implications for surgical planning and prediction of seizure-free outcomes in individual patients.

High resolution in-vivo diffusion imaging of the human hippocampus

The human hippocampus is a key target of many imaging studies given its capacity for neurogenesis, role in long term potentiation and memory, and nearly ubiquitous involvement in neurological and psychiatric conditions. Diffusion tensor imaging (DTI) has detected microstructural abnormalities of the human hippocampus associated with various disorders, but acquisitions have typically been limited to low spatial resolution protocols designed for whole brain (e.g. > 2 mm isotropic, >8 mm³ voxels), limiting regional specificity and worsening partial volume effects. The purpose here was to develop a simple DTI protocol using readily available standard single-shot EPI at 3T, capable of yielding much higher spatial resolution images (1 x 1 x 1 mm³) of the human hippocampus in a 'clinically feasible' scan time of ~6 min. A thin slab of twenty 1 mm slices oriented along the long axis of the hippocampus enabled efficient coverage and a shorter repetition time, allowing more diffusion weighted images (DWIs) per slice per unit time. In combination with this strategy, a low b value of 500 s/mm² was chosen to help overcome the very low SNR of a 1 x 1 x 1 mm³ EPI acquisition. 1 mm isotropic mean DWIs (averaged over 120-128 DWIs) showed excellent detail of the hippocampal architecture (e.g. morphology and digitations, sub-regions, stratum lacunosum moleculare - SLM) that was not readily visible on 2 mm isotropic diffusion images. Diffusion parameters within the hippocampus were consistent across subjects and fairly homogenous across sub-regions of the hippocampus (with the exception of the SLM and tail). However, it is expected that DTI parameters will be sensitive to microstructural changes associated with a number of clinical disorders (e.g. epilepsy, dementia) and that this practical, translatable approach for high resolution acquisition will facilitate localized detection of hippocampal pathology.

National Institutes of Health Stroke Scale Score is an Unreliable Predictor of Perfusion Deficits in Acute Stroke

Background

Perfusion-weighted magnetic resonance imaging is not routinely used to investigate stroke/transient ischemic attack. Many clinicians use perfusion-weighted magnetic resonance imaging selectively in patients with more severe neurological deficits, but optimal selection criteria have never been identified.

Aims and/or Hypothesis

We tested the hypothesis that a National Institutes of Health Stroke Scale score threshold can be used to predict the presence of perfusion-weighted magnetic resonance imaging deficits in patients with acute ischemic stroke/transient ischemic attack.

Methods

National Institutes of Health Stroke Scale scores were prospectively assessed in 131 acute stroke/transient ischemic attack patients followed by magnetic resonance imaging, including perfusion-weighted magnetic resonance imaging within 72 h of symptom onset. Patients were dichotomized based on the presence or absence of perfusion deficits using a threshold of Tmax (time to peak maps after the impulse response) delay ≥four-seconds and a hypoperfused tissue volume of ≥1 ml.

Results

Patients with perfusion deficits (77/131, 59%) had higher median (interquartile range) National Institutes of Health Stroke Scale scores (8 [12]) than those without perfusion deficits (3 [4], P < 0·001). A receiver operator characteristic analysis indicated poor to moderate sensitivity of National Institutes of Health Stroke Scale scores for predicting perfusion deficits (area under the curve = 0·787). A National Institutes of Health Stroke Scale score of ≥6 was associated with specificity of 85%, but sensitivity of only 69%. No National Institutes of Health Stroke Scale score threshold identified all cases of perfusion-weighted magnetic resonance imaging deficits with sensitivity >94%.

Conclusions

Although higher National Institutes of Health Stroke Scale scores are predictive of perfusion deficits, many patients with no clinically detectable signs have persisting cerebral blood flow changes. A National Institutes of Health Stroke Scale score threshold should therefore not be used to select patients for perfusion-weighted magnetic resonance imaging. Perfusion-weighted magnetic resonance imaging should be considered in all patients presenting with acute focal neurological deficits, even if these deficits are transient.