Mercedes P. Jacobson

Homocysteine and bone loss in epilepsy

Epidemiological studies reveal fracture incidence in epilepsy is twice that of the normal population. Much interest has been focused on Vitamin D, however, considering mixed results on non-enzyme inducing anti-epileptic drugs (AEDs) and bone mineral density (BMD) additional metabolic effects may be to blame. AEDs increase serum homocysteine (s-Hcy) by lowering blood folate levels. An association between elevated homocysteine, BMD and increased fracture incidence has been found in non-epilepsy populations. Additionally, folate and Vitamin B12 levels are independently related to bone mineral density in various non-epilepsy populations. This study supports previous research, which found elevated s-Hcy in subjects taking AEDs and that bone loss is related to the use of enzyme-inducing AEDs and changes in alkaline phosphatase. By one-way ANOVA, subjects on phenytoin monotherapy had significantly higher levels of s-Hcy than those on other AEDs (F=5.89, p=.016). Regression analyses revealed homocysteine, fracture history, length of years on AEDs, ethnicity were predictors of spine T scores. Weight and BMI were predictors of both BMD and DEXA T scores. Use of enzyme-inducing AEDs was a negative predictor of spine BMD and T scores, while phenytoin monotherapy was a positive predictor of spine BMD. Lamotrigine was found to be a negative predictor of spine T score. Ambulatory status, menopause and alcohol consumption were predictors of BMD but not T scores. In this study, persons with epilepsy who take nutritional supplementation have 25% lower s-Hcy levels than those who do not. Supplementation continues to be important in preventative epilepsy care.

Cardiovascular risk factors and homocysteine in epilepsy

Epidemiological studies have found the risk for heart disease and stroke are increased in persons with epilepsy. Anti-epileptic drugs (AEDs) have varying effects on serum lipids and homocysteine-an independent risk factor for coronary disease. The prevalence of cardiovascular risk factors (high cholesterol, hypertension, diabetes, obesity and smoking) and homocysteine were investigated in a multiethnic epilepsy population. Data included demographics, clinical factors, lab assessments and supplementation patterns. Mean age was 45 years (71 males and 94 females)-75 were African American, 27 Latino and 60 Caucasian. Fifty-two percent of participants had two or more cardiovascular risk factors when compared with rates for the general population of 28%. The Framingham risk score (FRS) assessment was also used to compare risk levels. Twenty-nine percent of men and 1% of women had a FRS indicating >5% level of risk, only 7% had a FRS>10%. Cardiovascular screening and primary preventative recommendations based on the American Heart Association and supplementation should be suggested for the adult epilepsy population when appropriate.

Self-efficacy, knowledge, health beliefs, quality of life, and stigma in relation to osteoprotective behaviors in epilepsy

It is well reported in the epilepsy literature that use of antiepileptic drugs (AEDs) leads to bone loss. Validated instruments were administered to assess knowledge, health behavior, quality of life, and stigma, to determine their effects on self-efficacy for osteoprotective and self-management behaviors. This adult epilepsy population had a mean age of 45, with 20 years of AED exposure. Fifty subjects were Caucasian and 44 were non-Caucasian. By one-way ANOVA, there were significant differences in self-efficacy based on ethnicity, medical assistance, status, and seizure frequency. Differences in knowledge based on ethnicity, education, and income were also noted. Regression analysis revealed that the factors that most predict self-efficacy for calcium, exercise, and self-management do not parallel each other. Age and ethnicity were predictive of self-efficacy for epilepsy self-management only. Medical management factors varied among the models. Overall quality of life was a positive predictor for both calcium and exercise self-efficacy.

Use of the Precaution Adoption Process Model to examine predictors of osteoprotective behavior in epilepsy

In the neurology literature it is well established that anti-epileptic drugs (AEDs) lead to bone loss (osteopenia and osteoporosis). Several large epidemiologic studies have found twice the fracture rate in persons with epilepsy compared to the non-epilepsy population. While an increasing level of awareness for preventative measures and screening by neurologists and primary care physicians are recommended, so far no one has attempted to address how knowledge related to calcium and exercise, health beliefs (based on the Health Belief Model) and self-efficacy (confidence in abilities) impact osteoprotective behaviors in epilepsy, based on the Precaution Adoption Process Model (PAPM). The seven-stage PAPM, unlike other health behavior theories where a person is either practicing or not practicing the behavior, conceptualizes behavior change as dynamic and occurring over time. Validated instruments were used to assess knowledge, health beliefs, self-efficacy and stages of the precaution adoption process for four osteoprotective behaviors. For dietary calcium; exercise knowledge and calcium self-efficacy predicted higher stages of precaution adoption. For calcium supplements; age perceived susceptibility for osteoporosis and perceived benefits of calcium predicted higher stages. Exercise adoption stage was most predicted by exercise knowledge and health motivation. For DEXA screening adoption; age and perceived susceptibility predicted higher stages. This study provides hints how persons with epilepsy could be influenced to move from the unaware/unengaged positions into to the stages of adoption and maintenance for osteoprotective behaviors.

Bone loss in epilepsy: Barriers to prevention, diagnosis, and treatment

In epilepsy populations there are multiple barriers to the prevention, diagnosis, and treatment of metabolic bone disease. This study explored issues that may be unique to inner-city care. We screened 101 consecutive patient charts from our clinic to identify barriers to optimal care. According to the Health Belief Model, when patients view a condition as not severe and their perceived susceptibility is low, they are unlikely to comply with screening and/or adopt preventative behaviors. Despite routine communications to primary care doctors regarding bone health, its importance is still overlooked. In addition, the managed care referral process makes it difficult for a specialist to prevent, screen, and treat a primary care condition with no symptoms. For optimal patient care and to stimulate prevention and screening compliance, a communication style adapted to the patients knowledge, beliefs, and attitudes toward antiepileptic drug-induced bone loss is crucial to help reduce barriers against change.

The TUH EEG CORPUS: A big data resource for automated EEG interpretation

The Neural Engineering Data Consortium (NEDC) is releasing its first major big data corpus - the Temple University Hospital EEG Corpus. This corpus consists of over 25,000 EEG studies, and includes a neurologists interpretation of the test, a brief patient medical history and demographic information about the patient such as gender and age. For the first time, there is a sufficient amount of data to support the application of state of the art machine learning algorithms. In this paper, we present pilot results of experiments on the prediction of some basic attributes of an EEG from the raw EEG signal data using a 3,762 session subset of the corpus. Standard machine learning approaches are shown to be capable of predicting commonly occurring events from simple features with high accuracy on closed-loop testing, and can deliver error rates below 50% on a 6-way open set classification problem. This is very promising performance since commercial technology fails on this data.

Osteoprotective knowledge in a multiethnic epilepsy population.

&NA; Antiepileptic drugs (AEDs) are known to cause bone loss. People with epilepsy have twice the fracture rate of nonepilepsy populations. Osteoprotective knowledge related to calcium and exercise has not been assessed in people with epilepsy. The Osteoporosis Knowledge Test (OKT), a validated, 24‐item test, was administered to 94 epilepsy patients (28 males and 66 females) to measure knowledge of risk factors for osteoporosis and strategies for prevention related to calcium and exercise. The mean age of participants was 45 years with an average AED exposure of 20 years. Fifty participants were Caucasian and 44 were non‐Caucasian. No significant differences related to age or gender for the OKT were found. One‐way analysis of variance (ANOVA) of ethnicity showed that non‐Caucasians had much lower calcium (F = 8.15, p = .005) and exercise (F = 7.71, p = .007) knowledge. The total mean OKT score was 11.71 (4.92), reflecting a correct response rate of 49%. In previous studies of nonepilepsy populations, the mean OKT score ranged from 7.83 to 21.8, with a correct response ranging from 32.9% to 90.8%. Independent t tests of the individual OKT questions revealed specific knowledge deficiencies in the areas of risk factors, exercise, and reasons for calcium supplementation for non‐Caucasians. Results of this study reveal that people with epilepsy, who are at greater risk for metabolic bone loss, have lower knowledge scores for calcium and exercise than nonepilepsy populations of various ages and genders. Culturally relevant epilepsy materials and programs may improve knowledge and adoption of preventative behavior.

Characteristics and acute outcomes of ICU patients with initial presentation of seizure

PURPOSE Seizures are a common cause of presentation to the emergency department (ED) but vary widely in severity and prognostic significance, with some cases requiring ICU management. Most evidence regarding seizure outcomes in the ICU comes from patients exclusively with status epilepticus (SE) or with new seizures detected after ICU admission. To aid in determining early prognosis of ICU patients with and without SE, we performed an analysis of patients initially presenting with any type of seizure and requiring ICU management. METHOD Analysis of hospital records of 247 consecutive patients presenting to the ED initially with seizure and directly admitted to the ICU between January 2010 and June 2013. The primary outcome was composite in-hospital death or discharge to hospice, and the secondary outcome was recurrent ICU seizures. RESULTS The primary outcome occurred in 7.7% of patients. Both early mechanical ventilation and an acute intracranial process on neuroimaging were associated with a poor outcome. About half of this cohort presented with SE. Although SE was associated with recurrent seizures in the ICU, the primary outcome was similar between patients presenting with and without SE. Patients with SE had greater rates of early intubation in the ED and were treated more aggressively with medication, whereas patients without SE had greater rates of first ever seizure, acute intracranial disease including intracranial hemorrhage, and neurosurgical intervention. CONCLUSION Patients presenting to the ED with and without SE requiring ICU admission may have similar acute outcomes, yet differ in risk factors and seizure etiologies.

Occipital dysembryoplastic neuroepithelial tumor presenting as adult-onset temporal epilepsy

Dysembryoplastic neuroepithelial tumor (DNET) is a benign brain tumor which commonly presents as childhood-onset temporal lobe epilepsy (TLE). We present a case of histologically proven DNET with a clinical presentation and scalp EEG suggestive of adult-onset TLE. MRI showed an occipital lesion. PET showed abnormal metabolism of the occipital lesion and the ipsilateral temporal lobe; raising concern for an abnormal functional network reorganization. Intracranial EEG showed interictal spikes and seizures originating from the occipital lesion with no seizures emanating from the temporal lobe. Occipital DNET due to their chronic nature can reorganize the network and mimic TLE.

The TUH EEG CORPUS

The Neural Engineering Data Consortium (NEDC) is releasing its first major big data corpus - the Temple University Hospital EEG Corpus. This corpus consists of over 25,000 EEG studies, and includes a neurologists interpretation of the test, a brief patient medical history and demographic information about the patient such as gender and age. For the first time, there is a sufficient amount of data to support the application of state of the art machine learning algorithms. In this paper, we present pilot results of experiments on the prediction of some basic attributes of an EEG from the raw EEG signal data using a 3,762 session subset of the corpus. Standard machine learning approaches are shown to be capable of predicting commonly occurring events from simple features with high accuracy on closed-loop testing, and can deliver error rates below 50% on a 6-way open set classification problem. This is very promising performance since commercial technology fails on this data.