New research was presented at AAN 2019, the American Academy of Neurology Annual Meeting, from May 4-10 in Philadelphia. The features below highlight some of the studies that emerged from the conference.
Connecting Treating Neurologists With Genomicists
Experience indicates that a main obstacle to routine genetic testing for neurology patients is the correlation of the thousands of genomics variants identified with the disease process in each unique patient. With few neurologists possessing the in-depth knowledge of genomics that this requires and testing laboratories not possessing in-depth knowledge of the individual patient, study investigators sought to introduce and assess a model that utilizes telemedicine consultations between a clinical genomicist and the treating neurologist to understand and apply genetic information to alter and improve treatment. Through the service, the patient and DNA sequences are presented through a telemedicine encounter, followed by a discussion regarding the meaning and utility of variants thought to be disease associated, with an emphasis on “actionable” findings that alter management. Among enrolled physicians and patients, variants of clinical interest that altered the physicians’ management were identified in 94% of cases, a diagnosis was provided in 53%, changes were recommended in therapy in 88%, additional testing was ordered in 41%, and outside referral was made in 24%.
—————————————————————-
Cognitive Reserve in MS During Cognitive Rehab
Previous research has shown that preservation of normal functional connectivity mitigates the association between focal white matter tract disruption and cognition in patients with multiple sclerosis (MS) and to relate to traditional proxies for cognitive reserve. To determine whether preserved functional connectivity moderates the relationship between white matter tract disruption and individual responsiveness to restorative cognitive rehabilitation in patients with MS, researchers recruited such patients for a 12-week restorative cognitive rehabilitation program. Participants underwent pre- and post-rehabilitation neuropsychological evaluations. Overall, participants experienced statistically significant increases in cognitive processing speed following rehabilitation. After accounting for age, gender, gray matter volume, disease duration, and T2-lesion volume, prediction of treatment efficacy was improved in a model also including white matter tract disruption in a localized network of four region-pairs centered on the precuneus in the left hemisphere.
—————————————————————-
Trends in Anticoagulation for Stroke Due to AF
Guidelines on the timing and anticoagulant choice for ischemic stroke due to atrial fibrillation (AF) provide imprecise recommendations. To assess current opinions of neurologists specializing in stroke, researchers presented case scenarios describing patients with acute ischemic stroke (AIS) of varying severity and hemorrhagic transformation (HT) incidence due to non-valvular paroxysmal AF to all US board-certified stroke neurologists through an Internet-based questionnaire. In patients with small AIS of less than one-third middle cerebral artery (MCA) territory, only 24% said they would start anticoagulation within 48 hours from stroke onset, whereas 80% anticoagulated by 7 days. With increased stroke severity of more than one-half MCA territory, only 29% chose to anticoagulate within 7 days, 48% waited 7-14 days, and 23% delayed past 14 days, with some requesting stability imaging before anticoagulating. Asymptomatic HT did not appreciably affect anticoagulation timing, but with symptomatic HT, 79% waited more than 14 days. Direct oral anticoagulants (DOACs) were the preferred anticoagulation strategy (62%), and low bleeding risk, affordability/cost, and availability of a specific reversal agent were the most common reasons cited (82%, 55%, 42% respectively). Aspirin was preferred by 57% in anticoagulation-ineligible patients.
—————————————————————-
Distinguishing Frontotemporal Dementia from Alzheimer’s
Prior studies suggest that self-conscious emotions may be disproportionately impaired among patients with behavioral variant frontotemporal dementia (bvFTD) and may help distinguish patients with bvFTD early in their disease from those with Alzheimer’s disease (AD) or health controls (HCs). For a study, people in each of these populations were administered a 36-item Embarassability Scale containing items of self-embarrassing situations and items of embarrassment for others. As a measure of sympathetic arousal, participants underwent continuous recordings of skin conductance levels (SCLs) while watching themselves on a 2-minute, pre-recorded video. In order to elicit embarrassability during the viewing video, participants were aware of being intensely watched and evaluated for their reactions. Patients with dementia also underwent (MRI with analysis for regions of interest. When compared with patients with AD and HCs, those with bvFTD had significantly less embarassability and significantly lower changes in skin conductance levels (compared with baseline) across the 2-minute viewing. Weak correlations were observed between skin conductance level changes and total and subscore embarrassability scores.
—————————————————————-
Monitoring PD Motor Symptoms With a Smartwatch
Data indicate that Parkinson’s disease (PD) motor manifestations vary greatly by the time of assessment in relation to the last does of PD medications. Wearable devices may allow for the tracking of these motor symptoms outside the clinic, providing useful information for appropriately managing treatment and monitoring disease progression. To test this hypothesis, study participants with PD performed a series of 13 tasks, simulating routine behaviors, while wearing a smartwatch device. Trained clinicians evaluated tremor and bradykinesia of both arms and rated moto symptoms using the MDS-UPDRS Part III assessment. The whole procedure was performed while both on and off medication. Machine learning models were trained to predict the symptom scores for each task based on the smartwatch accelerometry data. Two additional models were trained to predict the Part III subscore, based on symptom scores for the motor tasks, either actual or predicted. Correlations between actual and predicted Part III subscores were evaluated. Actual Part III subscores were moderately correlated with predicted subscores from clinician task scores and predicted subscores from accelerometry-predicted task scores. The most important tasks for accurate prediction were gross and fine upper body movements, though postural tasks also contributed.
