Stewart A Factor

Emory University, Atlanta, Georgia, United States

Are you Stewart A Factor?

Claim your profile

Publications (211)1209.13 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: A 12-month double-blind sham-surgery-controlled trial assessing adeno-associated virus type 2 (AAV2)-neurturin injected into the putamen bilaterally failed to meet its primary endpoint, but showed positive results for the primary endpoint in the subgroup of subjects followed for 18 months and for several secondary endpoints. Analysis of postmortem tissue suggested impaired axonal transport of neurturin from putamen to substantia nigra. In the present study, we tested the safety and efficacy of AAV2-neurturin delivered to putamen and substantia nigra. We performed a 15- to 24-month, multicenter, double-blind trial in patients with advanced Parkinson disease (PD) who were randomly assigned to receive bilateral AAV2-neurturin injected bilaterally into the substantia nigra (2.0 × 10(11) vector genomes) and putamen (1.0 × 10(12) vector genomes) or sham surgery. The primary endpoint was change from baseline to final visit performed at the time the last enrolled subject completed the 15-month evaluation in the motor subscore of the Unified Parkinson's Disease Rating Scale in the practically defined off state. Fifty-one patients were enrolled in the trial. There was no significant difference between groups in the primary endpoint (change from baseline: AAV2-neurturin, -7.0 ± 9.92; sham, -5.2 ± 10.01; p = 0.515) or in most secondary endpoints. Two subjects had cerebral hemorrhages with transient symptoms. No clinically meaningful adverse events were attributed to AAV2-neurturin. AAV2-neurturin delivery to the putamen and substantia nigra bilaterally in PD was not superior to sham surgery. The procedure was well tolerated, and there were no clinically significant adverse events related to AAV2-neurturin. Ann Neurol 2015. © 2015 American Neurological Association.
    Annals of Neurology 06/2015; DOI:10.1002/ana.24436 · 11.91 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Research suggests an association between global cognition and postural instability/gait disturbance (PIGD) in Parkinson disease (PD), but the relationship between specific cognitive domains and PIGD symptoms is not clear. This study examined the association of cognition (global and specific cognitive domains) with PIGD symptoms in a large, well-characterized sample of individuals with PD. Cognitive function was measured with a detailed neuropsychological assessment, including global cognition, executive function, memory, visuospatial function, and language. PIGD symptoms were measured using the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) Part III, Motor Examination subscale. Multiple linear regression analyses were performed to assess the relationship between cognition and PIGD symptoms with models adjusting for age, sex, education, enrollment site, disease duration, and motor symptom severity. The analysis included 783 participants, with mean (standard deviation) age of 67.3 (9.7) years and median (interquartile range) MDS-UPDRS Motor Subscale score of 26 (17, 35). Deficits in global cognition, executive function, memory, and phonemic fluency were associated with more severe PIGD symptoms. Deficits in executive function were associated with impairments in gait, freezing, and postural stability, while visuospatial impairments were associated only with more severe freezing, and poorer memory function was associated only with greater postural instability. While impairments in global cognition and aspects of executive functioning were associated with more severe PIGD symptoms, specific cognitive domains were differentially related to distinct PIGD components, suggesting the presence of multiple neural pathways contributing to associations between cognition and PIGD symptoms in persons with PD. Copyright © 2015 Elsevier Ltd. All rights reserved.
    Parkinsonism & Related Disorders 04/2015; 21(7). DOI:10.1016/j.parkreldis.2015.04.002 · 4.13 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Increasing evidence suggests that genetic factors play a role in the variability associated with cognitive performance in Parkinson's disease (PD). Mutations in the LRRK2 gene are the most common cause of monogenic PD; however, the cognitive profile of LRRK2-related PD is not well-characterized. A cohort of 1,447 PD patients enrolled in the PD Cognitive Genetics Consortium was screened for LRRK2 mutations and completed detailed cognitive testing. Associations between mutation carrier status and cognitive test scores were assessed using linear regression models. LRRK2 mutation carriers (n = 29) demonstrated better performance on the Mini Mental State Examination (P = 0.03) and the Letter-Number Sequencing Test (P = 0.005). A smaller proportion of LRRK2 carriers were demented (P = 0.03). Our cross-sectional study demonstrates better performance on certain cognitive tests, as well as lower rates of dementia in LRRK2-related PD. Future longitudinal studies are needed to determine whether LRRK2 mutation carriers exhibit slower cognitive decline. © 2015 International Parkinson and Movement Disorder Society. © 2015 International Parkinson and Movement Disorder Society.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Sleep disturbance and cognitive impairment are correlated in non-diseased populations, but their association in Parkinson's disease (PD) is uncertain. Prior studies employing measures of cognition and actigraphically-measured sleep have produced conflicting findings. Objective: In this descriptive study, we correlated measurements of sleep in PD patients derived from the gold-standard measurement, in-lab polysomnography, with an extensive battery of cognitive measures. We hypothesized that poorer sleep would be related to relatively more impaired cognition. Methods: Idiopathic PD patients (n = 34) completed a cognitive battery encompassing three broad domains (executive function, immediate memory and delayed memory), and underwent PSG for two nights. Scores for each domain from individual cognitive measures were converted to z-scores and then averaged to produce a composite score. We used second night PSG data and quantified measures of sleep architecture, sleep continuity, sleep apnea and nocturnal movement (periodic leg movements, PLMS). Results: Lower executive function was associated with higher PLMS after controlling for chronological age, mini-mental state examination scores, and UPDRS motor subscale scores. These results were independent from psychomotor speed. There was a marginally significant positive correlation between the proportion of time spent in REM and immediate recall ability. Measures of sleep continuity and sleep apnea were unrelated to cognition in these patients. Conclusions: PLMS, known to be a frequent feature of PSG-measured sleep in PD, may be an important correlate of impaired executive function in PD. Whether treating this disorder of sleep results in improvement in cognition remains to be determined.
    Journal of Parkinson's Disease 01/2015; 5(1). DOI:10.3233/JPD-140475 · 1.10 Impact Factor
  • H.A. Jinnah, Stewart A. Factor
    [Show abstract] [Hide abstract]
    ABSTRACT: The dystonias are a group of disorders characterized by excessive involuntary muscle contractions leading to abnormal postures and/or repetitive movements. A careful assessment of the clinical manifestations is helpful for identifying syndromic patterns that focus diagnostic testing on potential causes. If a cause is identified, specific etiology-based treatments may be available. In most cases, a specific cause cannot be identified, and treatments are based on symptoms. Treatment options include counseling, education, oral medications, botulinum toxin injections, and several surgical procedures. A substantial reduction in symptoms and improved quality of life is achieved in most patients by combining these options. Copyright © 2015 Elsevier Inc. All rights reserved.
    Neurologic Clinics 11/2014; 33(1). DOI:10.1016/j.ncl.2014.09.002 · 1.61 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Several case reports and small series have indicated that tardive dystonia is responsive to globus pallidus deep brain stimulation. Whether different subtypes or distributions of tardive dystonia are associated with different outcomes remains unknown.Methods We assessed the outcomes and temporal profile of improvement of eight tardive dystonia patients who underwent globus pallidus deep brain stimulation over the past six years through record review. Due to the retrospective nature of this study, it was not blinded or placebo controlled.ResultsConsistent with previous studies, deep brain stimulation improved the overall the Burke-Fahn-Marsden motor scores by 85.1±13.5%. The distributions with best responses in descending order were upper face, lower face, larynx/pharynx, limbs, trunk, and neck. Patients with prominent cervical dystonia demonstrated improvement in the Toronto Western Spasmodic Torticollis Rating Scale but improvements took several months. In four patients the effects of deep brain stimulation on improvement in Burke Fahn Marsden score was rapid, while in four cases there was partial rapid response of neck and trunk dystonia followed by was gradual resolution of residual symptoms over 48 months.Conclusion Our retrospective analysis shows excellent resolution of tardive dystonia after globus pallidus deep brain stimulation. We found instantaneous response, except with neck and trunk dystonia where partial recovery was followed by further resolution at slower rate. Such outcome is encouraging for using deep brain stimulation in treatment of tardive dystonia.
    Parkinsonism & Related Disorders 11/2014; 21(2). DOI:10.1016/j.parkreldis.2014.11.013 · 4.13 Impact Factor
  • PLoS Genetics 11/2014; 10(11):e1004774. DOI:10.1371/journal.pgen.1004774 · 8.17 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Hallucinations and delusions that complicate Parkinson's disease (PD) could lead to nursing home placement and are linked to increased mortality. Cognitive impairments are typically associated with the presence of hallucinations but there are no data regarding whether such a relationship exists with delusions. Objective: We hypothesized that hallucinations would be associated with executive and visuospatial disturbance. An exploratory examination of cognitive correlates of delusions was also completed to address the question of whether they differ from hallucinations. Methods: 144 PD subjects completed a neuropsychological battery to assess cognition and the SAPS to examine psychosis. Correlational analyses assessed associations between hallucinations and delusions with cognitive domains. Results: 48 subjects (33%) reported psychotic symptoms: 25 (17%) experienced hallucinations without delusions, 23 (16%) had symptoms dominated by delusions. Severity and/or number of hallucination subtypes were significantly correlated with lower scores in language, memory, attention, executive functioning, and visuospatial ability. Correlations with delusions were non-significant. Tests of differences in the size of the correlations between groups revealed a significant relationship between language and visuospatial performance with hallucinations. Conclusions: Cognitive correlates of hallucinations and delusions appear to be different in PD, suggesting distinct pathogenic mechanisms and possibly anatomical substrates. Hence, delusions may not share the same associations with dementia as hallucinations. Since this is a new finding, further studies will be needed to confirm our results.
    Journal of the Neurological Sciences 10/2014; 347(1-2). DOI:10.1016/j.jns.2014.10.033 · 2.26 Impact Factor
  • Neurology 10/2014; 83(15):1388-9. DOI:10.1212/01.wnl.0000455698.16732.0a · 8.30 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Freezing of gait (FOG) is major concern for Parkinson’s disease (PD) patients because it is a leading cause of falls and is associated with poor quality of life. The pathophysiology is unknown but it is hypothesized that it relates to cognitive abnormalities; particularly executive and visuospatial dysfunction. However, prior results have been discrepant. Pharmacologic subtypes of FOG include those that are responsive and unresponsive to levodopa. Objective To determine whether executive and visuospatial dysfunction are associated specifically with the levodopa unresponsive subtype of FOG. Methods 135 PD subjects completed a single assessment included FOG questionnaire, UPDRS motor scale, comprehensive cognitive battery and measure of hallucinations. Analyses compared unresponsive (n=16), responsive (n=20) and no FOG (n=99) subtypes. Results The unresponsive subtype had a significantly older age of onset of PD than the responsive group (p=.03) and had worse motor scores (p=.003) than the no FOG group. Longer disease duration was associated with the responsive group compared to the no FOG group (p=.002). The unresponsive FOG group had significantly poorer visuospatial ability (p=.001) and executive functioning (p=.02) than both the no and responsive FOG subgroups. These latter groups were not significantly different. The responsive FOG group was associated with the presence of hallucinations. Conclusion Aside from pharmacological differences, unresponsive FOG is associated with executive and visuospatial dysfunction implicating frontostriatal pathways while responsive FOG is associated with hallucinations suggesting involvement of posterior cortical regions. Further study and treatment of FOG should include appropriate subtype classification.
    Parkinsonism & Related Disorders 09/2014; 20(12). DOI:10.1016/j.parkreldis.2014.09.023 · 4.13 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction Generalized dystonia, both primary and secondary forms, and axial dystonias such as tardive dystonia, and idiopathic cervical dystonia are responsive to globus pallidus interna (GPi) DBS. There is a paucity of investigations probing the impact of DBS on adult-onset axial dystonia. We assessed the efficacy of GPi DBS in four patients with rare adult-onset axial dystonia. Methods Primary outcome measure was improvement in the motor component of the Burke-Fahn-Marsden (BFM) rating scale. Secondary outcome measures were quality of life as determined by the SF-36 questionnaire, time to achieve best possible benefit and DBS parameters that accounted for the best response. In patients with prominent concomitant cervical dystonia we also used the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS). Results GPi DBS improved BFM scores by 87.63 ± 11.46%. Improvement in total severity scale of TWSTRS was 71.5 ± 12.7%. Quality of life also remarkably improved as evidenced by 109.38 ± 82.97 and 7.05 ± 21.48% percent change in psychometrically-based physical component summary (PCS), and a mental component summary (MCS) score respectively. Conclusions GPi DBS is a very effective treatment for adult-onset axial dystonia. Considering its refractoriness to medical therapy and significant impact on quality of life DBS should be considered for this disorder.
    Parkinsonism & Related Disorders 09/2014; 20(11). DOI:10.1016/j.parkreldis.2014.09.005 · 4.13 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Cognitive impairment, including dementia, is common in Parkinson's disease (PD). The Mini-Mental State Examination (MMSE) has been recommended as a screening tool for Parkinson's disease dementia (PDD), with values below 26 indicative of possible dementia. Using a detailed neuropsychological battery, we examined the range of cognitive impairment in PD patients with an MMSE score of 26 or higher. In this multicenter, cross-sectional, observational study, we performed neuropsychological testing in a sample of 788 PD patients with MMSE scores of 26 or higher. Evaluation included tests of global cognition, executive function, language, memory, and visuospatial skills. A consensus panel reviewed results for 342 subjects and assigned a diagnosis of no cognitive impairment, mild cognitive impairment, or dementia. Sixty-seven percent of the 788 subjects performed 1.5 standard deviations below the normative mean on at least one test. On eight of the 15 tests, more than 20% of subjects scored 1.5 standard deviations or more below the normative mean. Greatest impairments were found on Hopkins Verbal Learning and Digit Symbol Coding tests. The sensitivity of the MMSE to detect dementia was 45% in a subset of participants who underwent clinical diagnostic procedures. A remarkably wide range of cognitive impairment can be found in PD patients with a relatively high score on the MMSE, including a level of cognitive impairment consistent with dementia. Given these findings, clinicians must be aware of the limitations of the MMSE in detecting cognitive impairment, including dementia, in PD. © 2014 International Parkinson and Movement Disorder Society
    Movement Disorders 09/2014; 29(10). DOI:10.1002/mds.25924 · 5.63 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cognitive impairment is a common and disabling problem in Parkinson disease (PD) that is not well understood and is difficult to treat. Identification of genetic variants that influence the rate of cognitive decline or pattern of early cognitive deficits in PD might provide a clearer understanding of the etiopathogenesis of this important nonmotor feature.
    JAMA Neurology 09/2014; 71(11). DOI:10.1001/jamaneurol.2014.1455 · 7.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Increased gut permeability, inflammation, and colonic α-synuclein pathology are present in early Parkinson's disease (PD) and have been proposed to contribute to PD pathogenesis. Peptidoglycan is a structural component of the bacterial cell wall. Peptidoglycan recognition proteins (PGRPs) maintain healthy gut microbial flora by regulating the immune response to both commensal and harmful bacteria. We tested the hypothesis that variants in genes that encode PGRPs are associated with PD risk. Participants in two independent case-control studies were genotyped for 30 single-nucleotide polymorphisms (SNPs) in the four PGLYRP genes. Using logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI) adjusted for potential confounding variables, we conducted analyses in each study, separately and pooled. One SNP failed the assay, and three had little to no variation. The ORs were similar in both study populations. In pooled analyses, three of seven PGLYRP2 SNPs (rs3813135, rs733731, rs892145), one of five PGLYRP3 SNPs (rs2987763), and six of nine PGLYRP4 SNPs (rs10888557, rs12063091, rs3006440, rs3006448, rs3006458, and rs3014864) were significantly associated with PD risk. Association was strongest for PGLYRP4 5'untranslated region (UTR) SNP rs10888557 (GG reference, CG OR 0.6 [95%CI 0.4-0.9], CC OR 0.15 [95%CI 0.04-0.6]; log-additive P-trend, 0.0004). Common variants in PGLYRP genes are associated with PD risk in two independent studies. These results require replication, but they are consistent with hypotheses of a causative role for the gut microbiota and gastrointestinal immune response in PD. © 2014 International Parkinson and Movement Disorder Society
    Movement Disorders 08/2014; 29(9). DOI:10.1002/mds.25895 · 5.63 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Quantitative MRI of neuromelanin (NM) containing structures (referred to as NM-MRI) in the brainstem, namely the locus coeruleus (LC) and substantia nigra (SN), may assist with the early detection of Parkinson's disease (PD) and Alzheimer's disease (AD) as well as differential diagnosis in the early disease stages. In this study, two gradient echo (GRE) sequences with magnetization transfer contrast (MTC) preparation pulses were developed to simultaneously image the LC and SN. This has been a challenge with NM-MRI techniques used in previous studies due to the relatively high specific absorption rate (SAR) induced by these techniques. In addition, a semi-automated quantitative analysis scheme was applied to estimate volumes and contrast-to-noise ratios (CNR) of the LC and SN based on segmentation of both structures. Compared to a T1-weighted turbo spin echo (TSE) sequence typically used for simultaneous imaging of the LC and SN, the two GRE-MTC sequences exhibited improved performance in terms of higher sensitivity (in CNR) in imaging the SN and lower SAR during the scans. A multiple-measurement protocol was adopted as well, so that motion degraded measurements could be removed and artifacts associated with motion corrected. The presented approach has demonstrated advantages in imaging acquisition (lower SAR and higher sensitivity), imaging pre-processing (with motion correction) and quantitative image analysis (segmentation-based estimation of volume and CNR) as compared with existing NM-MRI approaches. This approach has potential for detection and monitoring of neurodegeneration in LC and SN in disease states including AD and PD.
    Magnetic Resonance Imaging 07/2014; 32(10). DOI:10.1016/j.mri.2014.07.003 · 2.02 Impact Factor
  • Donald L Bliwise, Michael K Scullin, Stewart A Factor
    Journal of Neurology 06/2014; 261(7). DOI:10.1007/s00415-014-7392-x · 3.84 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Antidepressants have appeared to be more effective than placebo treatment in treating depressive syndromes in patients with Parkinson’s disease (PD). Objective To identify factors that predict improvement in depressive symptoms during antidepressant treatment in depressed PD patients. Methods A secondary analysis was performed on the dataset of the Randomized Placebo-controlled Study of Antidepressants in PD (SAD-PD), in which 76 patients received active treatment with either paroxetine or venlafaxine extended release (XR), and 39 patients received placebo treatment. Backward stepwise regression analyses were conducted with change in 24-item Hamilton Depression Rating Scale (HAMD-24) score between assessments at baseline and week 12 as the main outcome measure, and sex, age, baseline HAMD-24 score, Unified Parkinson’s Disease Rating Scale section III (UPDRS-III) score, Mini-Mental State Examination (MMSE), and the Clinical Anxiety Scale (CAS) as independent variables. Results In both the active treatment and placebo groups, higher baseline HAMD-24 score and lower UPDRS-III score were associated with greater reduction in HAMD-24 score. Higher anxiety scores predicted less response in the active treatment group. Higher MMSE scores predicted greater response only in the placebo treated group. Sex and age were no predictors of response. Conclusions Higher pre-treatment depression scores and lower pre-treatment anxiety scores are the two most important predictors for improvement during antidepressant treatment in depressed PD patients, which is in line with those found in treatment studies of depressed non-PD patients. Furthermore, our results indicate the requirement for different or more intensive treatment for depressed PD patients with more severe anxiety symptoms.
    Parkinsonism & Related Disorders 06/2014; 20(6). DOI:10.1016/j.parkreldis.2014.02.025 · 4.13 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The underlying etiology of parkinsonian anterocollis has been the subject of recent debate. The purpose of this study is to test the hypothesis that anterocollis in parkinsonian syndromes is associated with dystonia of the deep cervical flexors (longus colli and capitis). Eight patients with anterocollis, six in the setting of a Parkinsonism and two primary cervical dystonia control subjects with anterocollis underwent prospective structured clinical evaluations (interview, examination and rating scales), systematic electromyography of the cervical extensor musculature and 18 F-FDG PET/CT studies of cervical muscles to examine evidence of hypermetabolism or overactivity of deep cervical flexors. Subjects with parkinsonian anterocollis were found to have hypermetabolism of the extensor and sub-occipital muscles but not in the cervical flexors (superficial or deep). EMG abnormalities were observed in all evaluated patients, but only one patient was definitely myopathic. Meanwhile, both dystonia controls exhibited hypermetabolism of cervical flexors (including the longus colli). In conclusion, we were able to demonstrate hypermetabolism of superficial and deep cervical flexors with muscle 18 F-FDG PET/CT in dystonic anterocollis patients, but not in parkinsonian anterocollis patients. The hypermetabolic changes seen in parkinsonian anterocollis patients in posterior muscles may be compensatory. Alternative explanations for anterocollis include myopathy of the cervical extensors, or unbalanced rigidity of the cervical flexors, but this remains to be proven.
    Journal of the neurological sciences 05/2014; 340(1-2). DOI:10.1016/j.jns.2014.03.023 · 2.26 Impact Factor
  • Neurology 03/2014; 82(11):912-913. DOI:10.1212/WNL.0000000000000227 · 8.30 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Parkinson's disease (PD) is complex and heterogeneous. The numerous susceptibility loci that have been identified reaffirm the complexity of PD but do not fully explain it; e.g., it is not known if any given PD susceptibility gene is associated with all PD or a disease subtype. We also suspect that important disease genes may have escaped detection because of this heterogeneity. We used presence/absence of family history to subdivide the cases and performed genome-wide association studies (GWAS) in Sporadic-PD and Familial-PD separately. The aim was to uncover new genes and gain insight into the genetic architecture of PD. Employing GWAS on the NeuroGenetics Research Consortium (NGRC) dataset stratified by family history (1565 Sporadic-PD, 435 Familial-PD, 1986 controls), we identified a novel locus on chromosome 1p21 in Sporadic-PD (PNGRC = 4x10-8) and replicated the finding (PReplication = 6x10-3; PPooled = 4x10-10) in 1528 Sporadic-PD and 796 controls from the National Institutes of Neurologic Disease and Stroke (NINDS) Repository. This is the fifth PD locus to be mapped to the short of arm of chromosome 1. It is flanked by S1PR1 and OLFM3 genes, and is 200 kb from a multiple sclerosis susceptibility gene. The second aim of the study was to extend the stratified GWAS to the well-established PD genes. SNCA_ rs356220 was associated with both Sporadic-PD (OR = 1.37, P = 1x10-9) and Familial-PD (OR = 1.40, P = 2x10-5). HLA_rs3129882 was more strongly associated with Sporadic-PD (OR = 1.38, P = 5x10-10) than Familial-PD (OR = 1.12, P = 0.15). In the MAPT region, virtually every single nucleotide polymorphism (SNP) had a stronger effect-size and lower P-value in Familial-PD (peak P = 8x10-7) than in Sporadic-PD (peak P = 2x10-5). We discovered and replicated a new locus for Sporadic-PD which had escaped detection in un-stratified GWAS. This demonstrates that by stratifying on a key variable the power gained due to diminished heterogeneity can sometimes outweigh the power lost to reduced sample size. We also detected distinct patterns of disease associations for previously established PD susceptibility genes, which gives an insight to the genetic architecture of the disease and could aid in the selection of appropriate study population for future studies.
    BMC Genomics 02/2014; 15(1):118. DOI:10.1186/1471-2164-15-118 · 4.04 Impact Factor

Publication Stats

6k Citations
1,209.13 Total Impact Points


  • 2006–2015
    • Emory University
      • Department of Neurology
      Atlanta, Georgia, United States
    • Baylor College of Medicine
      Houston, Texas, United States
  • 2014
    • Wayne State University
      • Department of Neurology
      Detroit, Michigan, United States
    • Zoo Atlanta
      Atlanta, Georgia, United States
  • 2006–2008
    • University of Washington Seattle
      • • Department of Environmental and Occupational Health Sciences
      • • International Clinical Research Center (ICRC)
      Seattle, WA, United States
  • 2005–2008
    • Albany Stratton VA Medical Center
      Albany, New York, United States
    • Rhode Island Hospital
      Providence, Rhode Island, United States
  • 1989–2006
    • Albany Medical College
      • Department of Neurology
      Albany, New York, United States
  • 1999–2005
    • Parkinson's and Movement Disorders Center Of Maryland
      Maryland, United States
  • 2004
    • University of Florida
      • Department of Medicine
      Gainesville, FL, United States
  • 2002
    • Mayo Clinic - Scottsdale
      Scottsdale, Arizona, United States
  • 2001
    • University of Texas Southwestern Medical Center
      • Division of Neuro-oncology
      Dallas, TX, United States
  • 1989–2001
    • University of Miami Miller School of Medicine
      • Department of Neurology
      Miami, Florida, United States
  • 2000
    • Memorial Hospital of Rhode Island
      Pawtucket, Rhode Island, United States
  • 1987–2000
    • University of Miami
      • Department of Neurology
      كورال غيبلز، فلوريدا, Florida, United States
  • 1997
    • Harvard University
      Cambridge, Massachusetts, United States
    • University of Toronto
      • Division of Neurology
      Toronto, Ontario, Canada