November 2015
Volume 56, Issue 12
Free
Eye Movements, Strabismus, Amblyopia and Neuro-ophthalmology  |   November 2015
Skewness of Fractional Anisotropy Detects Decreased White Matter Integrity Resulting From Acute Optic Neuritis
Author Affiliations & Notes
  • Bradley G. Goodyear
    Department of Radiology, University of Calgary, Calgary, Alberta, Canada
    Deparment of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada
    Deparment of Psychiatry, University of Calgary, Calgary, Alberta, Canada
    Deparment of Electrical and Computer Engineering, University of Calgary, Calgary, Alberta, Canada
    Seaman Family MR Research Centre, University of Calgary, Calgary, Alberta, Canada
    Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
  • Nourhan M. H. Zayed
    Deparment of Electrical and Computer Engineering, University of Calgary, Calgary, Alberta, Canada
  • Filomeno Cortese
    Department of Radiology, University of Calgary, Calgary, Alberta, Canada
    Seaman Family MR Research Centre, University of Calgary, Calgary, Alberta, Canada
    Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
  • Jessie Trufyn
    Deparment of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada
  • Fiona Costello
    Deparment of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada
    Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
    Deparment of Surgery, University of Calgary, Calgary, Alberta, Canada
  • Correspondence: Bradley G. Goodyear, Seaman Family MR Research Centre, Foothills Medical Centre/University of Calgary, 1403, 29th Street NW, Calgary, AB, Canada T2N 2T9; goodyear@ucalgary.ca
Investigative Ophthalmology & Visual Science November 2015, Vol.56, 7597-7603. doi:10.1167/iovs.15-17335
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Bradley G. Goodyear, Nourhan M. H. Zayed, Filomeno Cortese, Jessie Trufyn, Fiona Costello; Skewness of Fractional Anisotropy Detects Decreased White Matter Integrity Resulting From Acute Optic Neuritis. Invest. Ophthalmol. Vis. Sci. 2015;56(12):7597-7603. doi: 10.1167/iovs.15-17335.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: Diffusion tensor imaging (DTI) has demonstrated optic nerve damage associated with optic neuritis (ON); however, the usefulness of mean fractional anisotropy (FA) specifically is varied in the literature. We wished to determine whether histogram analysis of FA better detects ON damage than mean FA.

Methods: The ON patients (n = 24) underwent DTI within 1 month of symptoms and then 6 months later (n = 21). Twelve control subjects participated in one session. Mean FA and axial (AD), radial (RD), and mean (MD) diffusivities were compared between ON and fellow eyes, control eyes, and sessions. Values were sorted into bins, and coefficients of skewness of FA, AD, RD, and MD were compared between ON and fellow eyes, control eyes, and sessions.

Results: Mean AD, RD, and MD of ON eyes were significantly reduced compared with fellow eyes (P < 0.04) within 1 month of symptoms, but did not differ at 6 months. Mean AD and RD of ON eyes were significantly lower than control eyes (P < 0.05). No differences were observed for mean FA. The coefficient of skewness of FA was significantly different between ON eyes and fellow eyes (P = 0.03) and control eyes (P = 0.04) within 1 month of symptoms, but did not differ at 6 months. No differences were observed for AD, RD, and MD.

Conclusions: Skewness of FA can detect white matter damage associated with ON and its recovery, which may further inform us of how DTI can measure white matter injury and repair.

Optic neuritis (ON) is an inflammatory injury of the optic nerve and frequently presents with pain and subacute onset vision loss.1 Optic neuritis may occur in isolation or in the setting of multiple sclerosis (MS). In fact, approximately 20% of MS patients experience ON as the first clinical manifestation of their disease.2 Resolution of visual symptoms typically occurs within weeks, although visual recovery may continue for up to 1 year.1 Yet, despite typically achieving high contrast visual acuity equivalent of 20/20 (Snellen) acuity in the postacute phase, many ON patients report persistent visual dysfunction, including problems with motion and depth perception and heat-induced vision loss.1,2 
The mechanism of optic nerve injury is thought to involve destruction of the insulating myelin sheath surrounding the nerve and the consequent death of the denuded axons.35 The anatomical structure and function of the optic nerves makes them easily assessable using quantitative clinical and electrophysiologic measures, relative to other central nervous system (CNS) pathways.69 Moreover, the ability to use highly quantifiable measures of afferent visual pathway structure and function makes ON a good model to understand mechanisms of disease injury contributing to disability in MS. 
Magnetic resonance imaging (MRI) permits a direct assessment of optic nerve structure in the presence of ON.1012 The spatial extent of abnormalities observed in contrast-enhanced T1-weighted MR images of the optic nerve exhibit a significant correlation with severity of visual impairment at the time of the acute inflammatory event,13,14 and nerve involvement at the site of the optic canal is associated with impaired color vision. However, the location and extent of MRI lesions do not necessarily predict recovery.15 Mean optic nerve area at 1-year after injury, as determined by fluid-attenuated inversion recovery images, is also not associated with visual outcome.5 
Optic nerve anatomy also lends itself to investigation using diffusion tensor imaging (DTI),1618 an MRI technique that provides information about the random motion of water molecules, referred to as “diffusion.” Due to the hydrophilic nature of the myelin sheath and the orientation of optic nerve fibers, the preferential direction of water diffusion is along the nerve. Increases in water diffusion (quantified as mean diffusivity [MD]17 or the apparent diffusion coefficient [ADC]16) have been reported at 1-year after injury in cases of persistent visual acuity deficits. Axial diffusivity (AD) (i.e., diffusion along the nerve) is reduced in the ON eye during the acute phase (less than 1 month)1921 and appears to be predictive of visual outcome.21,22 Thus, DTI may represent a structural marker of axonal integrity, which when applied to ON as a putative model, may provide important insights regarding mechanisms of white matter damage and repair in CNS inflammatory disorders. 
Another DTI metric, fractional anisotropy (FA), is a composite measure of the diffusivities that quantifies the directional preference of water diffusion on a scale of 0 to 1, where 0 indicates complete nonpreferential directionality and 1 indicates one direction only. The geometry and orientation of the optic nerve makes FA particularly useful in theory for evaluating its structural integrity. Patients who do not experience full recovery 1 year after acute ON exhibit reduced FA along the ON-affected nerve,1618,23 possibly indicating persistent disruption of the myelin sheath or permanent axonal loss. Although FA provides a simple quantitative assessment of water motion direction, its utility to date seems limited to remote or more severe cases of axonal disruption in which longitudinal changes are relatively robust. This may be due to the fact that FA is not a linear measure and has a limited range of 0 to 1. That is, a change in FA from 0.55 to 0.65, for example, is not equivalent to a change from 0.85 to 0.95. Fractional anisotropy also only quantifies the directional preference of diffusion and not the magnitude of diffusion itself. Thus, FA values obtained from the optic nerve can only shift up or down within a distribution bounded by 0 and 1. Although diffusivities are also bounded in theory (zero indicates no diffusion and free water represents the upper limit), white matter anatomy and associated disruptions do not typically approach these limits. Hence, the analysis of mean FA may be inadequate for the assessment of optic nerve integrity and the detection of white matter disease in the context of progression or recovery from an inflammatory injury. An alternative approach is histogram analysis, which examines how values are distributed about the data mean or mode. For example, Pearson's coefficient of skewness quantifies the symmetry of how values are distributed about the mean or mode. It is therefore plausible that changes in FA can be captured by changes in the skewness of the data. 
Several DTI studies have used histogram analysis in investigations of the visual pathway excluding the optic nerve,24,25 as well as other white matter regions in the CNS.26,27 None, however, have investigated the optic nerve directly in the presence of acute ON injury or inflammation. Therefore, the primary objective of the current study was to determine whether histogram analysis of FA (i.e., skewness) is better than mean FA at distinguishing ON eyes from fellow eyes and healthy control eyes at the time of the acute inflammatory event, as well as differentiate acute and chronic phases of ON in affected eyes. 
Methods
Study Design and Participants
This prospective cohort study included 24 consecutively enrolled patients (mean age, 33.8 years; range, 22–49 years; 21 females and 3 males) referred for evaluation of acute ON to the University of Calgary Neuro-Ophthalmology and MS Clinics. The diagnosis of ON was based on the presence of eye pain at onset, a history of acute or subacute decreased monocular visual acuity, evidence of a relative afferent pupillary defect in the affected eye, color desaturation in the affected eye, a visual field defect consistent with the diagnosis of ON, and a fundus examination showing a normal appearing optic nerve or mild optic disc edema at the time of symptom onset. Patients had acute ON either as a clinically isolated syndrome (CIS; n = 16) or in the context of clinically or radiologically confirmed relapsing remitting MS (RRMS; n = 8).2831 Patients were excluded from the study if they had a history of prior ON in either eye or another reason for vision loss including amblyopia, glaucoma, or another optic neuropathy. Patients suspected of having ON related to systemic connective tissue disorders, neuromyelitis optica, infection, or nondemyelinating CNS inflammatory syndromes were also excluded from the study. Contraindications to MR imaging were also a basis for exclusion. Patients were not randomized to a particular treatment arm in this study, and the decision to forego therapy or use high-dose corticosteroid therapy was left to the discretion of the patient's primary treating physician. 
Twelve age-matched healthy volunteers (mean age, 33.1 years; range, 24–46 years; 11 females and 1 male) with no known neurologic or ophthalmic disorders also participated in one MRI session as control subjects. The ethics board of the institution approved this study, which followed the tenets of the Declaration of Helsinki, and all participants provided informed written consent prior to their participation. 
Ophthalmic Testing
All patients and control subjects underwent repeat neuro-ophthamic testing, which included best-corrected high-contrast letter acuity using Early Treatment Diabetic Retinopathy Study charts at 3.2 m (Lighthouse Low-Vision Products, Long Island City, NY, USA) and Snellen visual acuity. The right eye was tested first, followed by the left eye. Visual acuity scores were recorded as the 4-m logarithm of the minimum angle of resolution (logMAR), where each 0.1 increase in logMAR acuity represents a one-line decline in visual acuity.32 Count fingers and hand motion vision were recorded and converted to 1.6 and 2.0 logMAR, respectively. Visual field testing was performed with Humphrey perimetry (Carl Zeiss Meditec, Dublin, CA, USA), using the Central 30-2, full-threshold strategy. Slit lamp biomicroscopy and dilated ophthalmoscopy were also performed at each visit. 
Diffusion Tensor Imaging and Postprocessing
All patients underwent an MRI session within 1 month of the onset of symptoms and were symptomatic at the time of imaging (defined as “acute”). Twenty-one patients returned for a follow-up session approximately 6 months later (mean, 5.5 ± 0.8 months; range, 4–7 months). Control subjects underwent one MRI session. These individuals were recruited throughout the entire study period. 
All images were collected using a 3-T MR scanner equipped with an eight-channel phased-array radiofrequency head coil (Signa VHi; GE Healthcare, Waukesha, WI, USA). Foam padding was used to minimize head movements, and participants were instructed to lie still and close their eyes. Sagittal anatomical images were collected first to identify the optic nerve. An iterative nonlinear shimming sequence (GE Healthcare) was used to minimize the heterogeneity of the magnetic field around the optic nerves. For DTI, a single-shot spin-echo echo planar imaging diffusion sequence was used (with fat suppression), with the following parameters: 11 diffusion directions, b = 850 s/mm2, one T2-weighted base image (i.e., the b0 image), field of view = 240 × 240 mm, 3-mm slice thickness, and two averages. Images were collected in a plane perpendicular to the optic nerve such that the nerve was imaged in cross sections over several slices (TR/TE = 8000/84.3 ms, 1.87 × 1.87 × 3 mm, 24 slices, 208-second scan time). Prior to image reconstruction, DTI images were zero-padded such that the resulting resolution was 0.94 × 0.94 mm. Whole-head T2-weighted imaging was performed to visually inspect the afferent visual pathway for the presence of sclerotic lesions. 
All image processing was performed without any knowledge of patient clinical characteristics or disease status. Using the FMRIB's Diffusion Toolbox v2.0 (http://www.fmrib.ox.ac.uk/fsl/fdt/index.html), images of FA, MD, AD, and RD were generated. Using the drawing tool of FSL (FSLView v3.1.8), the optic nerve was manually segmented with the aid of the b0 image, which does not possess diffusion information. An example is shown in Figure 1. This method has been shown previously to possess good intra- and interrater reliability for accurate segmentation of the optic nerve.20 For the first 16 patients scanned, the reproducibility of the optic nerve segmentation was assessed between two independent raters, as implemented in a previous DTI study of the optic nerve.21 For each participant (patients and control subjects), the mean of FA, MD, RD, and AD was calculated for each optic nerve. For each of FA, MD, RD, and AD, the values for all voxels were obtained, the minimum and maximum values were determined, and all voxels were grouped into 24 equal-width bins to obtain a histogram distribution, from which Pearson's coefficient of skewness about the histogram mode was calculated as (mean – mode)/standard deviation. 
Figure 1
 
T2-weighted (b0) images of the optic nerves. The images are perpendicular cross-sectional images through the optic nerves. Images to the left show the original images, and the images to the right show the optic nerve segmentation. Red indicates segmented right optic nerve, and green indicates segmented left optic nerve.
Figure 1
 
T2-weighted (b0) images of the optic nerves. The images are perpendicular cross-sectional images through the optic nerves. Images to the left show the original images, and the images to the right show the optic nerve segmentation. Red indicates segmented right optic nerve, and green indicates segmented left optic nerve.
Data Analysis
For each DTI measurement and each eye, the mean and coefficient of skewness from acute ON patients were each compared with controls using an independent samples t-test. This analysis was also applied to data collected at 6 months. 
For data collected during acute ON, a Spearman rank correlation analyses was performed between visual acuity of the ON eye and each of the mean and the coefficient of skewness of each DTI measurement. For both the mean and the coefficient of skewness, a paired samples t-test was used to compare ON eyes with fellow eyes. Paired samples t-tests were performed again for data collected at 6 months. All statistical tests were corrected for multiple comparisons using the family-wise error rate. P < 0.05 was considered as statistically significant, and values less than 0.1 were considered to demonstrate a strong trend toward significance. 
Results
Clinical Characteristics
From the first imaging session, the data of three ON patients were excluded due to excessive head motion during scanning, leaving data from 21 patients for analysis. The data from one ON patient were excluded from the 6-month scanning session, leaving data from 20 patients for analysis. As a result, 17 patients had data from both imaging sessions. All control participant data sets were analyzed. Patient demographics and clinical characteristics are summarized in the Table. During acute ON, patients exhibited reduced visual acuity of the ON eye, which as a group, improved to normal by 6 months (Table). All control subjects exhibited normal visual acuity. Visual inspection of T2-weighted images did not reveal any discernible lesions along the afferent visual pathway in any of our patients. 
Table
 
Participant Characteristics
Table
 
Participant Characteristics
DTI Findings: Mean Measurements Over the Nerve
Interrater reproducibility of optic nerve segmentation was high. The mean difference between raters in the voxels deemed to be in the optic nerve was 9.9% (SD, 6.4%). As a result, the difference in FA between raters was 1.2% (SD, 0.8%) and the difference in MD was 1.2% (SD, 1.1%). This is in good agreement with past studies using manual segmentation of the optic nerve in DTI images.21 
At the acute phase, in comparison with control eyes, there was a significant reduction in mean AD [t(31) = 2.08, P = 0.046] and RD [t(31) = 2.05, P = 0.049] in ON eyes (Fig. 2), as well as a strong trend for MD [t(31) = 1.78, P = 0.084]. There were no differences in DTI measures between fellow eyes and controls eyes acutely or after at 6 months. During the acute phase, in comparison with the fellow eye, there was a significant reduction in MD [t(20) = 2.61, P = 0.017], AD [t(20) = 2.41, P = 0.026] and RD [t(20) = 2.22, P = 0.038] in ON eyes (Fig. 2), whereas there were no differences at 6 months. No significant differences were observed for mean FA in any of these comparisons. Spearman rank correlation analysis demonstrated a strong trend between visual acuity of the ON eye, acutely, and each of MD [rs = −0.422, t(21) = −2.03, P = 0.057], AD [rs = −0.426, t(21) = −2.05, P = 0.054], and RD [rs = −0.425, t(21) = −2.04, P = 0.055; Fig. 3]. There was no trend for an association between visual acuity and FA in ON eyes. 
Figure 2
 
Diffusion measurements for patients (ON and fellow eyes) and controls. At the time of acute ON, mean diffusivity (all of MD, AD, and RD) of the ON eye was significantly less (P < 0.04) than the fellow eye. Mean AD and RD of the ON eye were significantly less than control eyes (P < 0.05). All measures of mean diffusivity increased significantly (P < 0.05) over the 6 months of recovery. No significant differences were observed for mean FA.
Figure 2
 
Diffusion measurements for patients (ON and fellow eyes) and controls. At the time of acute ON, mean diffusivity (all of MD, AD, and RD) of the ON eye was significantly less (P < 0.04) than the fellow eye. Mean AD and RD of the ON eye were significantly less than control eyes (P < 0.05). All measures of mean diffusivity increased significantly (P < 0.05) over the 6 months of recovery. No significant differences were observed for mean FA.
Figure 3
 
Diffusion measurements of ON eyes in relation with visual acuity at the time of the acute ON event. There is a strong trend (0.05 < P < 0.10) between diffusivity and visual acuity (as determined by Spearman ranked correlation analysis).
Figure 3
 
Diffusion measurements of ON eyes in relation with visual acuity at the time of the acute ON event. There is a strong trend (0.05 < P < 0.10) between diffusivity and visual acuity (as determined by Spearman ranked correlation analysis).
Compared with the acute phase, MD [t(16) = 2.59, P = 0.020], AD [t(16) = 2.42, P = 0.028], and RD [t(16) = 2.21, P = 0.042] significantly increased by 6 months in ON eyes (see also Fig. 2). No difference between the acute phase and 6 months was observed for FA in ON eyes, and no differences were observed in the fellow eyes of ON patients. 
DTI Findings: Skewness
At the acute phase, in comparison with control eyes, there was a significant difference in the coefficient of skewness for FA [t(31) = 2.14, P = 0.041; Fig. 4]. No differences were observed for the coefficient of skewness for MD, RD, or AD. There were no differences between fellow eyes and controls eyes and no differences at 6 months. In the acute phase, in comparison with the fellow eye, there was a significant difference in the coefficient of skewness for FA [t(20) = 2.73, P = 0.028] of the ON eye (Fig. 4). There were no differences at 6 months. No significant differences were observed for MD, RD, and AD in any of these comparisons. Spearman rank correlation analysis did not reveal any associations between skewness and visual acuity of the acute ON eye. 
Figure 4
 
Left: Pearson's coefficient of skewness of FA for patients (ON and fellow eyes) and controls. FA skewness was significantly different between acute ON eyes and fellow eyes (P = 0.028), and between acute ON eyes and control eyes (P = 0.041). The FA skewness of the ON eye changed significantly over the 6 months of recovery (P = 0.047). No other significant differences were observed. Right: The FA histograms of the percentage of optic nerve image voxels, relative to distribution mode (i.e., bin 0). The difference in skewness of the histogram can be observed between acute ON eyes and fellow eyes, and this difference in skewness is resolved by 6 months.
Figure 4
 
Left: Pearson's coefficient of skewness of FA for patients (ON and fellow eyes) and controls. FA skewness was significantly different between acute ON eyes and fellow eyes (P = 0.028), and between acute ON eyes and control eyes (P = 0.041). The FA skewness of the ON eye changed significantly over the 6 months of recovery (P = 0.047). No other significant differences were observed. Right: The FA histograms of the percentage of optic nerve image voxels, relative to distribution mode (i.e., bin 0). The difference in skewness of the histogram can be observed between acute ON eyes and fellow eyes, and this difference in skewness is resolved by 6 months.
Compared with the acute phase, the coefficient of skewness for FA [t(16) = 2.16, P = 0.047] significantly changed by 6 months in the ON eye, matching that of the fellow and control eyes (Fig. 2). No differences were observed for the coefficients of skewness for MD, AD, and RD, and no differences were observed in the fellow eye. 
DTI Measurements: Acute Management With High-Dose Corticosteroids for ON
Four patients received the equivalent of 1000 mg intravenous methylprednisolone for 5 days at the time of the acute ON event. Post hoc analyses of these patients as a separate group did not reveal any differences from the remaining patients, with respect to visual acuity scores or DTI measurements. 
Discussion
In this study, we examined DTI indicators of optic nerve integrity in consecutively enrolled ON patients, who were followed prospectively from the time of the acute inflammatory event to 6 months later. In this cohort, we observed reduced diffusivity of the ON nerve at the time of the acute inflammatory optic nerve injury, relative to fellow eyes, in agreement with previous studies.2022 Also consistent with previous studies,21,22 we observed that diffusivity returned to normal levels with recovery, and there was a near significant association between reduction in diffusivity and visual acuity in ON eyes during the acute phase. Our findings thus confirm the clinical utility of mean diffusivity of the entire optic nerve for the diagnosis and assessment of visual dysfunction caused by ON. 
The utility of FA to date has been primarily limited to ON patients with persistent optic nerve injury.17,23 In these cases, injury volume and severity are potentially sufficient for detection with mean FA over the entire nerve. Our findings suggest that mean FA of the entire optic nerve is not a good indicator of injury, at least for consecutively enrolled ON patients with varying degrees of visual dysfunction. Because FA is a composite measure of the diffusivities, the underlying physiologic mechanisms of a change in FA are likely the same as for changes in diffusivity. Thus, it is reasonable to expect that FA should also detect optic nerve injury. However, if in the ON eye, all the diffusivities are reduced relative to the fellow eye (as is the case for our data and the data of others) and if the decrease is the same percentage-wise for all diffusivities, then FA does not actually change. Hence, changes in FA are not only dependent on changes in the diffusivities, but also changes in the diffusivities relative to each other. This could lead to increases in FA in some cases and decreases in others. Simple calculation of a population mean may thus be insufficient to detect meaningful changes in FA in ON patients. We proposed histogram analysis of FA as means to address this issue. We demonstrated there is a redistribution of FA values within the optic nerve. Thus, histogram analysis (i.e., skewness) of FA appears to be a sensitive technique to detect the presence of acute ON injury in patient cohorts compiled without selection bias. 
Most ON cohorts are heterogenous by nature, including both CIS and MS patients. Indeed, some of our patients had been diagnosed with RRMS prior to the onset of ON. In these cases, it is possible that subclinical lesions along the afferent visual pathway, including the optic chiasm, the lateral geniculate nuclei, the optic radiations, or white matter tracts connecting other visual cortical areas, may impact visual acuity, both at the time of the acute ON event and after recovery. None of our patients had any discernible lesions along the afferent visual pathway. Therefore, to the best of our knowledge, only acute ON injury contributed to the visual disturbance experience by our patients. 
Our ON patients were not randomized to a particular treatment regime, and the decision to use high-dose corticosteroid therapy was left to the primary treating physician in this study. Our analysis, taking into account the patients that received intravenous methylprednisolone, did not alter our main findings. Hence, it is unlikely that steroid treatment impacted our findings in general, albeit the small sample size in this study limits any conclusions that can be drawn in this regard. 
Our values of FA for the optic nerve were low in comparison to previous studies.1618,23 This may be due to the somewhat lower spatial resolution of our DTI images. As a result, there is likely a combination of optic nerve and surrounding CSF in some image voxels. The agreement between our results and those of previous studies in terms of how diffusivity changes in the presence of acute ON suggests, however, that this partial voluming effect simply led to a systematic shift to lower values of FA (given the nondirectional preference of water diffusion in the CSF). Thus, this did not impact our ability to detect differences between ON eyes and fellow eyes and control eyes. One potential solution for our data would be to impose an intensity threshold to ensure regions of interest are within the nerve. However, this could lead to a selection bias toward higher FA values and could potentially miss areas of injury with decreased FA. Future studies at higher spatial resolution could also potentially spatially localize the image voxels contributing to the redistribution of FA values and may thus localize injury. 
In summary, in agreement with previous studies, DTI measurements of mean diffusivity robustly detect the presence of ON at the time of the acute event, and the normalization of diffusivity over the months following is associated with good visual recovery. Our novel finding that histogram analysis (skewness) detects a shift in the distribution of FA at the time of the acute event potentially offers a new means to assess acute injury due to ON. Our findings, in turn, may offer new insights into the potential role of DTI in tracking more global effects of injury and repair in CNS inflammatory disorders. 
Acknowledgments
Supported by the Multiple Sclerosis Society of Canada. 
Disclosure: B.G. Goodyear, None; N.M.H. Zayed, None; F. Cortese, None; J. Trufyn, None; F. Costello, None 
References
Setsompop K, Gagoski BA, Polimeni JR, Witzel T, Wedeen VJ, Wald LL. Blipped-controlled aliasing in parallel imaging for simultaneous multislice echo planar imaging with reduced g-factor penalty. Magn Reson Med. 2012 ; 67: 1210–1224.
Miller D, Barkhof F, Montalban X, Thompson A, Filippi M. Clinically isolated syndromes suggestive of multiple sclerosis part I: natural history, pathogenesis, diagnosis, and prognosis. Lancet Neurol. 2005 ; 4: 281–288.
Hood DC, Zhang X, Greenstein VC, et al. An interocular comparison of the multifocal VEP: a possible technique for detecting local damage to the optic nerve. Invest Ophthalmol Vis Sci. 2000 ; 41: 1580–1587.
Hickman SJ, Toosy AT, Miszkiel KA, et al. Visual recovery following acute optic neuritis—a clinical, electrophysiological and magnetic resonance imaging study. J Neurol. 2004 ; 251: 996–1005.
Hickman SJ, Toosy AT, Jones SJ, et al. A serial MRI study following optic nerve mean area in acute optic neuritis. Brain. 2004 ; 127: 2498–2505.
Weber AL, Klufas R, Pless M. Imaging evaluation of the optic nerve and visual pathway including cranial nerves affecting the visual pathway. Neuroimaging Clin North Am. 1996 ; 6: 143–177.
Pro MJ, Pons ME, Liebmann JM, et al. Imaging of the optic disc and retinal nerve fiber layer in acute optic neuritis. J Neurol Sci. 2006 ; 250: 114–119.
Smith MM, Strottmann JM. Imaging of the optic nerve and visual pathways. Semin Ultrasound CT MR. 2001 ; 22: 473–487.
Rocca MA, Hickman SJ, Bo L, et al. Imaging the optic nerve in multiple sclerosis. Mult Scler. 2005 ; 11: 537–541.
Miller DH, Johnson G, McDonald WI, et al. Detection of optic nerve lesions in optic neuritis with magnetic resonance imaging. Lancet. 1986 ; 1: 1490–1491.
Miller DH, Mac Manus DG, Bartlett PA, Kapoor R, Morrissey SP, Moseley IF. Detection of optic nerve lesions in optic neuritis using frequency-selective fat-saturation sequences. Neuroradiology. 1993 ; 35: 156–158.
Miller DH, Newton MR, van der Poel JC, et al. Magnetic resonance imaging of the optic nerve in optic neuritis. Neurology. 1988 ; 38: 175–179.
Kupersmith MJ, Alban T, Zeiffer B, Lefton D. Contrast-enhanced MRI in acute optic neuritis: relationship to visual performance. Brain. 2002 ; 125: 812–822.
Kupersmith MJ, Gal RL, Beck RW, Xing D, Miller N. Visual function at baseline and 1 month in acute optic neuritis: predictors of visual outcome. Neurology. 2007 ; 69: 508–514.
Kolappan M, Henderson AP, Jenkins TM, et al. Assessing structure and function of the afferent visual pathway in multiple sclerosis and associated optic neuritis. J Neurol. 2009 ; 256: 305–319.
Hickman SJ, Wheeler-Kingshott CA, Jones SJ, et al. Optic nerve diffusion measurement from diffusion-weighted imaging in optic neuritis. AJNR Am J Neuroradiol. 2005 ; 26: 951–956.
Trip SA, Wheeler-Kingshott C, Jones SJ, et al. Optic nerve diffusion tensor imaging in optic neuritis. Neuroimage. 2006 ; 30: 498–505.
Wheeler-Kingshott CA, Trip SA, Symms MR, Parker GJ, Barker GJ, Miller DH. In vivo diffusion tensor imaging of the human optic nerve: pilot study in normal controls. Magn Reson Med. 2006 ; 56: 446–451.
Bermel RA, Fox RJ. Picturing injury and recovery with diffusion tensor imaging: the eyes have it. Neurology. 2009 ; 72: 584–585.
Naismith RT, Xu J, Tutlam NT, et al. Disability in optic neuritis correlates with diffusion tensor-derived directional diffusivities. Neurology. 2009 ; 72: 589–594.
van der Walt A, Kolbe SC, Wang YE, et al. Optic nerve diffusion tensor imaging after acute optic neuritis predicts axonal and visual outcomes. PLoS One. 2013 ; 8: e83825.
Naismith RT, Xu J, Tutlam NT, et al. Diffusion tensor imaging in acute optic neuropathies: predictor of clinical outcomes. Arch Neurol. 2012 ; 69: 65–71.
Kolbe S, Chapman C, Nguyen T, et al. Optic nerve diffusion changes and atrophy jointly predict visual dysfunction after optic neuritis. Neuroimage. 2009 ; 45: 679–686.
El-Rafei A, Engelhorn T, Warntges S, Dorfler A, Hornegger J, Michelson G. Glaucoma classification based on visual pathway analysis using diffusion tensor imaging. Magn Reson Med. 2013; 31: 1081–1091.
Yu CS, Lin FC, Li KC, et al. Diffusion tensor imaging in the assessment of normal-appearing brain tissue damage in relapsing neuromyelitis optica. AJNR Am J Neuroradiol. 2006 ; 27: 1009–1015.
Dreha-Kulaczewski SF, Brockmann K, Henneke M, et al. Assessment of myelination in hypomyelinating disorders by quantitative MRI. J MRI. 2012 ; 36: 1329–1338.
Zhou Y, Qun X, Qin LD, Qian LJ, Cao WW, Xu JR. A primary study of diffusion tensor imaging-based histogram analysis in vascular cognitive impairment with no dementia. Clin Neurol Neurosurg. 2011 ; 113: 92–97.
Poser CM, Paty DW, Scheinberg L, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol. 1983 ; 13: 227–231.
Lublin FD, Reingold SC. Defining the clinical course of multiple sclerosis: results of an international survey. National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology. 1996 ; 46: 907–911.
Polman CH, Reingold SC, Edan G, et al. Diagnostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria.” Ann Neurol. 2005 ; 58: 840–846.
Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011 ; 69: 292–302.
Henderson AP, Trip SA, Schlottmann PG, et al. An investigation of the retinal nerve fibre layer in progressive multiple sclerosis using optical coherence tomography. Brain. 2008 ; 131: 277–287.
Figure 1
 
T2-weighted (b0) images of the optic nerves. The images are perpendicular cross-sectional images through the optic nerves. Images to the left show the original images, and the images to the right show the optic nerve segmentation. Red indicates segmented right optic nerve, and green indicates segmented left optic nerve.
Figure 1
 
T2-weighted (b0) images of the optic nerves. The images are perpendicular cross-sectional images through the optic nerves. Images to the left show the original images, and the images to the right show the optic nerve segmentation. Red indicates segmented right optic nerve, and green indicates segmented left optic nerve.
Figure 2
 
Diffusion measurements for patients (ON and fellow eyes) and controls. At the time of acute ON, mean diffusivity (all of MD, AD, and RD) of the ON eye was significantly less (P < 0.04) than the fellow eye. Mean AD and RD of the ON eye were significantly less than control eyes (P < 0.05). All measures of mean diffusivity increased significantly (P < 0.05) over the 6 months of recovery. No significant differences were observed for mean FA.
Figure 2
 
Diffusion measurements for patients (ON and fellow eyes) and controls. At the time of acute ON, mean diffusivity (all of MD, AD, and RD) of the ON eye was significantly less (P < 0.04) than the fellow eye. Mean AD and RD of the ON eye were significantly less than control eyes (P < 0.05). All measures of mean diffusivity increased significantly (P < 0.05) over the 6 months of recovery. No significant differences were observed for mean FA.
Figure 3
 
Diffusion measurements of ON eyes in relation with visual acuity at the time of the acute ON event. There is a strong trend (0.05 < P < 0.10) between diffusivity and visual acuity (as determined by Spearman ranked correlation analysis).
Figure 3
 
Diffusion measurements of ON eyes in relation with visual acuity at the time of the acute ON event. There is a strong trend (0.05 < P < 0.10) between diffusivity and visual acuity (as determined by Spearman ranked correlation analysis).
Figure 4
 
Left: Pearson's coefficient of skewness of FA for patients (ON and fellow eyes) and controls. FA skewness was significantly different between acute ON eyes and fellow eyes (P = 0.028), and between acute ON eyes and control eyes (P = 0.041). The FA skewness of the ON eye changed significantly over the 6 months of recovery (P = 0.047). No other significant differences were observed. Right: The FA histograms of the percentage of optic nerve image voxels, relative to distribution mode (i.e., bin 0). The difference in skewness of the histogram can be observed between acute ON eyes and fellow eyes, and this difference in skewness is resolved by 6 months.
Figure 4
 
Left: Pearson's coefficient of skewness of FA for patients (ON and fellow eyes) and controls. FA skewness was significantly different between acute ON eyes and fellow eyes (P = 0.028), and between acute ON eyes and control eyes (P = 0.041). The FA skewness of the ON eye changed significantly over the 6 months of recovery (P = 0.047). No other significant differences were observed. Right: The FA histograms of the percentage of optic nerve image voxels, relative to distribution mode (i.e., bin 0). The difference in skewness of the histogram can be observed between acute ON eyes and fellow eyes, and this difference in skewness is resolved by 6 months.
Table
 
Participant Characteristics
Table
 
Participant Characteristics
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×