Jennifer L. Cox

Retinal nerve fiber layer thickness is associated with brain MRI outcomes in multiple sclerosis

Multiple sclerosis is characterized by the dual pathological processes of inflammation and neurodegeneration. Conventional MRI techniques are considered the best tools for assessing and monitoring lesion burden and inflammation but are limited in their ability to assess axonal loss. Optical coherence tomography (OCT) is a simple high-resolution technique that uses near infrared light to quantify the thickness of the retinal nerve fiber layer (RNFL), which contains only non-myelinated axons. RNFL thickness (RNFLT) was measured using OCT on thirty consecutive MS patients (60 eyes). Eighteen patients underwent quantitative MRI analysis including T1- and T2-lesion volumes (LV), normalized brain volume (NBV), normalized cortical, white and gray matter volumes (NCV, NWMV, and NGMV), and mean whole brain diffusivity (MD). There was a strong association between NBV and average RNFL thickness (p<0.001, partial rp=0.77). The T2-LV and NWMV were significantly associated with average RNFL thickness (p=0.002, partial rp= -0.76 and p=0.005, partial rp=0.68, respectively) and there were trends toward association with T1-LV (p=0.041) and NGMV (p=0.067). There was negative correlation between average RNFL thickness (average of both eyes) and disability as assessed by EDSS (p=0.02). The results support potential usefulness of OCT for MS patient monitoring and research applications.

Extent of cerebellum, subcortical and cortical atrophy in patients with MS: A case-control study

Cortical and subcortical atrophy occurs in multiple sclerosis (MS) and relates to clinical outcomes. FreeSurfer, a voxel-based automated software for brain reconstruction was used to investigate the extent of subcortical and cortical atrophy in 71 MS and 17 clinically isolated syndrome (CIS) patients, and 38 normal controls (NC), and to relate group differences to disease type and severity. Segmentation was performed on 3D SPGR T1-weighted MRI 1.5T images. Region-specific subcortical tissue volumes were calculated in mm(3) and cortical thickness in mm. Logistic regression and general linear model analyses, adjusted for age and intracranial volume, examined differences between NC, MS and CIS patients and disease subtypes. The MS group was characterized by significantly lower volumes of thalamus (left and right p<0.0001), left inferior lateral ventricle, third ventricle (p<0.0001), ventral diencephalon, pallidum and putamen bilaterally, as well as of right accumbens and brainstem with corresponding bilateral increase in volumes of lateral ventricles (p<0.01). Focal cortical atrophy areas in the thalamus, inferior parietal lobule of left hemisphere and in right precuneus were also significant in the MS sample. Versus CIS patients, RR or progressive MS patients showed significantly lower volumes of subcortical regions and cortical thinning. Hippocampal atrophy appeared only in advanced disease stages. Cerebellum WM volumes were significantly lower in MS and CIS patients vs. NC. Subcortical and cortical atrophy correlated with higher disability as measured by EDSS. This study confirmed selective deep gray matter atrophy (mostly thalamic), revealed cerebellum WM atrophy from the earliest clinical stages, and showed that cortical thinning advances with disease progression.

Magnetic resonance imaging characteristics of children and adults with paediatric-onset multiple sclerosis

The purpose of this study was to compare the clinical and quantitative magnetic resonance imaging metrics of paediatric-onset multiple sclerosis to adult-onset multiple sclerosis. It was a prospective comparison of clinical and magnetic resonance imaging characteristics of two paediatric onset multiple sclerosis and two adult onset multiple sclerosis groups that were matched for disease duration. The paediatric-onset-C group consisted of children with paediatric-onset multiple sclerosis with mean disease duration of 2.7 years, whereas the paediatric onset-A group consisted of adults with mean disease duration of 20 years. The adult onset multiple sclerosis-1 and adult onset multiple sclerosis-2 groups were matched to the paediatric onset-C and paediatric onset-A groups. The brain magnetic resonance imaging measures included: T(1)-, T(2)- and gadolinium contrast-enhancing volumes and the T(2)-lesion volume relative magnetization transfer ratio, global and tissue specific white and grey matter brain atrophy and normal appearing grey and white matter magnetization transfer ratio. Regression analyses were employed for magnetic resonance imaging measures. The paediatric onset multiple sclerosis-C (n = 17) and adult onset multiple sclerosis-1 (n = 81) groups had mean disease duration values of 2.7 +/- standard deviation 2.0 and 2.6 +/- 1.1 years, respectively. The paediatric onset multiple sclerosis-A group (n = 33) and adult onset multiple sclerosis-2 group (n = 300) had mean disease durations of 20 +/- standard deviation 10.9 and 20 +/- 9.3 years, respectively. In regression analysis, the T(2)- lesion volume of the paediatric onset multiple sclerosis-C and adult onset multiple sclerosis-1 groups were similar but there was a trend toward higher T(1)- lesion volume (P = 0.028) in the paediatric onset group. The brain parenchymal fraction and grey matter fraction in the paediatric-onset multiple sclerosis-C group were higher than those for the adult onset multiple sclerosis-1 group (both P < 0.001). The frequency of progressive multiple sclerosis in the paediatric onset multiple sclerosis-A group (27.3%) trended lower (odds ratio = 0.43, P = 0.042) than that in the adult onset multiple sclerosis-2 group (46.3%). The Expanded Disability Status Scale (median; inter-quartile range) in the paediatric onset multiple sclerosis-A group (2.25; 2.5) trended lower (P = 0.058) compared with the adult onset multiple sclerosis-2 group (3.5; 4.0). There was a trend toward lower magnetization transfer ratio values in T(2)-lesions, normal appearing grey matter and normal appearing white matter and higher grey matter fraction in the paediatric onset multiple sclerosis-A group compared with the adult onset multiple sclerosis-2 group. There was no evidence for differences on T(2)-lesion volume, T(1)-lesion volume, brain parenchymal fraction or white matter fraction. Paediatric-onset multiple sclerosis is characterized by a significant disease burden both early and later in the disease course. Despite this, disability is slower to accrue in paediatric onset multiple sclerosis than adult onset multiple sclerosis.

Relationship of optic nerve and brain conventional and non-conventional MRI measures and retinal nerve fiber layer thickness, as assessed by OCT and GDx: A pilot study

BACKGROUND Measurement of retinal nerve fiber layer (RNFL) thickness in multiple sclerosis (MS) is gaining increasing attention. OBJECTIVES To explore the relationship between RNFL thickness as measured by optical coherence tomography (OCT) and scanning laser polarimetry with variable corneal compensation (GDx), and conventional and non-conventional optic nerve and brain MRI measures. METHODS Twelve relapsing-remitting (RR) MS patients (12 affected and 12 unaffected eyes) and 4 age- and sex-matched normal controls (NC) (8 unaffected eyes) were enrolled. Four MS patients had a history of bilateral optic neuritis (ON), four had a history of unilateral ON, and 4 had no history of ON. Optic nerve MRI measurements included the length of T2 lesions, measurement of optic nerve atrophy, magnetization transfer ratio (MTR) and diffusion tensor imaging (DTI) measures. Optic nerve atrophy was measured by a novel method with high reproducibility. Brain MRI measurements included T1 and T2 lesion volumes (LVs) and their relative MTRs, and tissue class specific atrophy, MTR and DTI measures. Measures of RNFL were evaluated with OCT and GDx. We also evaluated both high and low contrast letter acuities (LCLA) in order to determine the relationship between vision, MRI metrics, and retinal structural architecture. RESULTS LCLA, RNFL-OCT and optic nerve radius measures showed more robust differences between NC and MS patients, and between MS patients with affected and unaffected eyes. T2-LV and T1-LV, as well as gray matter atrophy, DTI and MTR measures were related to LCLA and RNFL thickness. Unique additive variance regression models showed that both brain and optic nerve MRI measures independently accounted for about 50% of the variance in LCLA and RNFL thickness. In reverse models, about 20% of the additional independent variance was explained by optic nerve or brain MRI metrics. CONCLUSIONS Measurement of RNFL thickness and radius of the optic nerve should be preferred to the other optic nerve MRI measures in clinical studies. Whole brain lesion and GM measures are predictive of impaired visual function with corresponding structural concomitants.

Evolution of different MRI measures in patients with active relapsing-remitting multiple sclerosis over 2 and 5 years: a case–control study

Background: There is growing evidence for the concept of multiple sclerosis (MS) as an inflammatory neurodegenerative disease, with a different pattern of atrophy evolution in grey matter (GM) and white matter (WM) tissue compartments. Objective: We aimed to investigate the evolution of different MRI measures in early relapsing-remitting patients with MS and in normal controls (NCs) over 2 years. We also evaluated the progression of these MRI measures in a subset of patients who were followed for up to 5 years. Methods: Included in this study were 147 patients who participated in the combination ASA (Avonex Steroids Azathioprine) study and completed full treatment, clinical and MRI assessment at 0, 12 and 24 months. A subgroup of 66 patients was followed for 36 months, 51 patients for 48 months and 43 patients for 60 months. Mean age at baseline was 30.7 years, mean disease duration was 5.5 years, mean EDSS was 1.8 and mean annualised relapse rate before study entry was 1.7. MRI scans were performed on a 1.5T scanner every 2 months for the first 2 years and thereafter once yearly for up to 5 years. In addition to the MS group, 27 NCs were examined at months 0, 12 and 24 using the same MRI protocol. Percentage brain volume change (PBVC), GM volume (GMV), WM volume (WMV) and peripheral grey volume (PGV) were measured annually using SIENA/X software. T2-hyperintense lesion volume (LV), lateral ventricle volume (LVV) and third ventricle width (3VW) were also assessed annually. Results: Over the period of 0–24 months, patients with MS lost significantly more GMV (−2.6% vs −0.72%, p<0.001), PGV (−2.4% vs −1.03%, p<0.001) and PBVC (−1.2% vs −0.22%, p<0.001), and increased in LVV (+16.6% vs +0.55%, p<0.003) and 3VW (+9.3% vs 0%, p = 0.003), when compared with NCs. Within-person change in MRI measures for patients with MS over 5 years was −4.2% for PBVC, −6.2% for GMV, −5.8% for PGV, −0.5% for WMV (all p<0.001), +68.7 for LVV (p<0.001), +4% for 3VW (p<0.001) and +42% for T2-LV (p<0.001). Conclusions: Our study confirmed a different pattern of GM, WM and central atrophy progression over 2 years between patients with MS and NCs. The study showed a different evolution of tissue compartment atrophy measures in patients with MS, with faster decline in cortical and deep GM regions, as well as periventricular WM regions, over a 5-year period.

Neuroimaging in multiple sclerosis.

Conventional magnetic resonance imaging (MRI) has routinely been used to improve the accuracy of multiple sclerosis (MS) diagnosis and prognosis. Metrics derived from conventional MRI are now routinely used to detect therapeutic effects and extend clinical observations. However, conventional MRI measures, such as the use of lesion volume and count of gadolinium-enhancing and T2 lesions, have insufficient sensitivity and specificity to reveal the true degree of pathological changes occurring in MS. They cannot distinguish between inflammation, edema, demyelination, Wallerian degeneration, and axonal loss. In addition, they do not show a reliable correlation with clinical measures of disability and do not provide a complete assessment of therapeutic outcomes. Recent neuropathologic studies of typical chronic MS brains reveal macroscopic demyelination in cortical and deep gray matter (GM) that cannot be detected by currently available MRI techniques. Therefore, there is a pressing need for the development of newer MRI techniques to detect these lesions. Newer metrics of MRI analysis, including T1-weighted hypointense lesions, central nervous system atrophy measures, magnetization transfer imaging, magnetic resonance spectroscopy, and diffusion tensor imaging, are able to capture a more global picture of the range of tissue alterations caused by inflammation and neurodegeneration. At this time, they provide the only proof--albeit indirect--that important occult pathology is occurring in the GM. However, evidence is increasing that these nonconventional MRI measures correlate better with both existing and developing neurological impairment and disability when compared to conventional metrics.

Randomized study of interferon beta-1a, low-dose azathioprine,and low-dose corticosteroids in multiple sclerosis

Background Studies evaluating interferon beta (IFNβ) for multiple sclerosis (MS) showed only partial efficacy. In many patients, IFNβ does not halt relapses or disability progression. One strategy to potentially enhance efficacy is to combine IFNβ with classical immunosuppressive agents, such as azathioprine (AZA) or corticosteroids, commonly used for other autoimmune disorders. Objective The Avonex–Steroids–Azathioprine study was placebo-controlled trial and evaluated efficacy of IFNβ-1a alone and combined with low-dose AZA alone or low-dose AZA and low-dose corticosteroids as initial therapy. Methods A total of 181 patients with relapsing–remitting MS (RRMS) were randomized to receive IFNβ-1a 30 μg intramuscularly (IM) once weekly, IFNβ-1a 30 μg IM once weekly plus AZA 50 mg orally once daily, or IFNβ-1a 30 μg IM once weekly plus AZA 50 mg orally once daily plus prednisone 10 mg orally every other day. The primary end point was annualized relapse rate (ARR) at 2 years. Patients were eligible for enrollment in a 3-year extension. Results At 2 years, adjusted ARR was 1.05 for IFNβ-1a, 0.91 for IFNβ-1a plus AZA, and 0.73 for combination. The cumulative probability of sustained disability progression was 16.8% for IFNβ-1a, 20.7% for IFNβ-1a plus AZA, and 17.5% for combination. There were no statistically significant differences among groups for either measure at 2 and 5 years. Percent T2 lesion volume change at 2 years was significantly lower for combination (+14.5%) versus IFNβ-1a alone (+30.3%, P < 0.05). Groups had similar safety profiles. Conclusion In IFNβ-naïve patients with early active RRMS, combination treatment did not show superiority over IFNβ-1a monotherapy.

Gray matter atrophy and disability progression in patients with early relapsing-remitting multiple sclerosis A 5-year longitudinal study

We assessed the relationship between gray matter (GM) and white matter (WM) atrophy and clinical status in early relapsing-remitting multiple sclerosis (MS) patients over 5 years. A group of 181 patients who participated in the ASA (Avonex-Steroid-Azathioprine) study and had complete clinical and MRI assessments over 2 and 5 years was investigated. One hundred seventy (170) patients completed the 12-month follow-up, 147 the 24-month, 98 the 36-month, 65 the 48-month and 47 the 60-month. Changes in GM (GMV), WM (WMV) and peripheral GM (PGV) volumes, whole brain volume (percentage brain volume change PBVC), lateral ventricle volume (LVV), third ventricle width (3VW) and T2-lesion volume (T2-LV) were measured. Patients were assigned according to their clinical status to one of two groups: the Stable group, and the Reached Confirmed Sustained Progression (RCSP) group (24-week interval). At 0-6 months PBVC and GMV, at 0-12 months PBVC, GMV and T2-LV, at 0-24 months PBVC and GMV, at 0-36 months PBVC, GMV and T2-LV, and at 0-48 PBVC predicted the differences between the RCSP and Stable groups. PBVC and LVV showed the strongest ability to differentiate patients who presented 0 or >or=3 relapses in the Stable group. Decline in PBVC and GMV were predictive markers of disability deterioration. Correlation of T2-LV with clinical status was weaker and decreased over time. Higher number of relapses was associated with faster decline in whole brain volume.

The place of conventional MRI and newly emerging MRI techniques in monitoring different aspects of treatment outcome

Magnetic resonance imaging (MRI) is the most important paraclinical measure for assessing and monitoring the pathologic changes implicated in the onset and progression of multiple sclerosis (MS). Conventional MRI sequences, such as T1-weighted gadolinium (Gd) enhanced and spin-echo T2-weighted imaging, only provide an incomplete picture of the degree of inflammation and underlying neurodegenerative changes in this disease. Two- and three-dimensional fluid-attenuated inversion recovery and double inversion recovery sequences allow better identification of cortical, periventricular and infratentorial lesions. Ultra-high field strength MRI has the potential to detect subpial cortical and deep gray matter lesions. Unenhanced T1-weighted imaging can reveal hypointense black holes, a measure of chronic neurodegeneration. Magnetization transfer imaging (MTI) is increasingly used to characterize the evolution of MS lesions and normal-appearing brain tissue. Evidence suggests that the dynamics of magnetization transfer changes correlate with the extent of demyelination and remyelination. Magnetic resonance spectroscopy, which provides details on tissue biochemistry, metabolism, and function, also has the capacity to reveal neuroprotective mechanisms. By measuring the motion of water, diffusion imaging can provide information about the orientation, size, and geometry of tissue damage in white and gray matter. These advanced non-conventional MRI techniques relate better to clinical impairment, disease progression, accumulation of disability, and have the potential to detect neuroprotective effects of treatment. Although detecting the status of neuronal integrity using MRI techniques continues to improve, a “gold standard” model remains to be established.

Gray matter correlations of cognition in incident Parkinson's disease

We aimed to investigate whether mild cognitive impairment (MCI) in Parkinsons disease (PD) is characterized by region‐specific gray matter (GM) atrophy and to explore correlations between GM and cognition in PD. Magnetic resonance images of 42 newly diagnosed PD patients (of which 11 had MCI) and 37 normal controls were analyzed using voxel‐based morphometry. Analyses comparing groups showed no regional atrophy, and in patients there were no significant correlations between cognitive domain test performance and GM loss. In conclusion, GM atrophy does not seem to be a major feature of cognitive dysfunction in incident PD.