You are here
The evaluation of MRI diffusion values of active demyelinating lesions in multiple sclerosis
Multiple Sclerosis and Related Disorders, Volume 10, November 2016, Pages 97–102
Gadolinium (Gd) enhancement of lesions is the main radiologic marker for detection of activity in Multiple Sclerosis (MS). This study compares Diffusion weighted imaging (DWI) characteristics and enhancement to determine whether DWI can be used as an alternative to Gd administration.
A retrospective study of 72 patients who had MRI with Gd and DWI. Visual assessment and comparison of the Apparent Diffusion Coefficient (ADC) values on Gd+ lesions, all lesions showing restricted diffusion, 2 Gd− lesions and 1 area of normal-appearing white matter (NAWM) in each MRI were performed.
DWI values were measured on 275 T2 lesions, 68 Gd+ and 207 Gd− lesions, as well as 104 NAWM. 34 Gd+ lesions showed restricted diffusion. The median ADC-minimum of Gd+ lesions was significantly lower than NAWM and even lower than Gd− lesions. Most DWI restricted lesions were also Gd+(specificity≥94%), however many Gd+ lesions did not show visually detectable restriction in DWI (sensitivity≤34%). The median ADC-minimum of symptomatic lesions was lower than asymptomatic lesions.
While Gd+ lesions have lower ADC-minimum, visual DWI assessment cannot replace Gd administration for identifying active lesions. Gd+ lesions showing restricted diffusion are clinically important as they are more likely associated with neurological symptoms.
- Diffusivity is heterogeneous in Gd+ MS lesions compared to Gd− lesions.
- Visually restrict lesions in ADC map often enhance with Gd on T1-weighted images.
- Visually restrict lesions in ADC map is a weak surrogate marker for enhancement.
- Gd− enhanced study cannot be replaced by DWI/ADC.
- Lower ADC values of lesions did correlate more tightly with the symptomatic lesions.
Keywords: Multiple sclerosis, Diffusion magnetic resonance imaging, Gadolinium, Magnetic resonance imaging, Restricted diffusion, Detection of new MRI activity.
1. Background and purpose
Conventional MRI sequences including T2/FLAIR hyperintensities and gadolinium (Gd) administration is an established paraclinical tool in diagnostic criteria of multiple sclerosis (MS) (Polman et al., 2011). MRI findings have also been widely used as surrogate markers for the monitoring of treatment efficacy in patients (Freedman et al., 2013). The importance of new activity in the course of MS was indicated in the recent revision of International Advisory Committee on Clinical Trials of MS (Lublin et al., 2014).
The gold standard in the detection of active demyelinating lesions over the course of the disease is the focal enhancement in a T1–weighted MRI after Gd injection. Serial MRI studies have shown that Gd− enhancement occurs in almost all new lesions in patients with relapsing remitting (RR) or secondary progressive (SP) forms of MS and can be sometimes detected even before the onset of clinical symptoms. Most lesions enhanced for less than 1 month and no longer than 6 months (Miller et al., 1988a, 1988b). The number of enhancing lesions increases shortly before and during clinical relapses and clinical relapses predict subsequent MRI activity (Molyneux et al., 1998). The majority of enhancing lesions however are asymptomatic (Miller et al., 1998) and are commonly seen even when there are no clinical suggestions of disease activity. The detection of a new T2 lesion, although more sensitive to detect interval activity, can be tricky in a clinical setting to monitor activity (i.e. due to long interval between serial scanning, different MRI technique, positioning or the presence of confluent lesions). It is certainly more time consuming than detecting Gd+ lesions and requires prior similarly acquired images for comparison (Sormani et al., 2013).
There are some concerns regarding using Gd− based contrast agents (GBCAs), which may justify considering an alternative MRI sequence as a substitute for a contrast enhanced MRI (CE MRI). GBCAs are sometimes, although rarely, contraindicated in situations such as in patients with acute or chronic severe renal insufficiency (glomerular filtration rate below 30 mL/min/1.73 m2) who are at increased risk for developing nephrogenic systemic fibrosis, and in patients with an allergy to GBCAs. More recently it was noted that Gd gets deposited in the body and parts of the brain and the long term effects of these deposits are not known (McDonald et al., 2015). There are not enough evidence-based studies regarding the safety of GBCAs during fetal development, and therefore these agents should be administered in pregnancy only in exceptional cases in which their usage is considered critical and the potential benefits justify the potential risk to the unborn fetus (Kanal et al, 1990, Manual on Contrast Media v9, 2014, and Thomsen et al, 2013). Lastly, the use of Gd increases the cost of MRI by some estimates of about 30% (MRI Private Costs - Scans and Rates, 2014), involving not only the cost of the GBCA but the time and personnel involved to infuse it and monitor for reactions.
Diffusion-weighted Imaging (DWI) is now routinely being used as part of the scanning protocol in MS patients and has been proposed as an adjunct for screening disease activity in MS. In some centers, restriction of diffusion in lesions has been proposed as a substitute for contrast-enhanced T1-weighted imaging to detect active or new lesions. Conflicting results have been obtained when comparing enhancing and non-enhancing lesions with respect to their appearance on DWI; some studies have reported higher apparent diffusion coefficient or mean diffusivity in non-enhancing compared with enhancing lesions (Roychowdhury et al, 2000 and Werring et al, 1999) whereas other studies of large numbers of patients and lesions have reported no significant difference between Gd+ and Gd− lesions (Filippi et al., 2000). The purpose of this study was to investigate the relationship of DWI metrics to the status of Gd− enhancement of demyelinating lesions in MS patients.
2. Patients and methods
2.1. Patients and study design
Patients were recruited via a clinical monitoring software (IMED, version 6.1) (iMed, 2014) that collects clinical and paraclinical data on more than 5000 outpatients attending the Ottawa Hospital MS Clinic since 1999. Our hospital research ethics board has approved this study and waived individual patients’ consent (OHSN-REB 20140268-01H). We screened 466 patients who attended the MS Clinic between 2007 and 2014 since the MRI scanner was updated in 2007. All patients met current international criteria for a diagnosis of MS or clinically isolated syndrome (CIS) by an experienced neurologist with the exclusion of other possible alternative diseases (such as acute disseminated encephalomyelitis or neuromyelitis optica). Only patients who had undergone MRI with Gd enhanced T1-weighted sequence and DWI protocols with available imaging in our PACS (Radiology Information System with Integrated Solutions, 2015) were included in the study. A review of the patients’ charts was performed and Expanded Disability Status Scale (EDSS) score (Kurtzke, 1983) at the time of MRI study and any clinical relapse 6 months before to 1 month after MRI studies were recorded.
2.2. MRI sequences and imaging analysis
The MRI studies were performed on either 1.5 or 3T Siemens MR Scanners. The brain MRI sequences included sagittal and axial T1, axial T2 fat suppressed, and T2-fluid-attenuated inversion recovery (T2-FLAIR) in the axial plane. DWI was acquired with a single-shot echo planar sequence in three orthogonal directions with diffusion gradients b-value of 0, 500 and 1000 s/mm2. Apparent diffusion coefficient (ADC) maps were automatically generated. The DWI was performed prior to administration of the Gd with a slice thickness of 5 mm. The ADC values are a quantitative measure of diffusivity and will have lower values in areas of restricted diffusion. These areas will appear hypointense (dark) in ADC map. Gd enhanced T1-weighted images were acquired in the axial and coronal planes after intravenous injection of 0.1 mmol/kg of Gd after approximately 5 min of injection.
The images were evaluated to detect Gd+ lesions and lesions showing restricted diffusion in DWI (hyperintense in DWI and hypointense in the ADC map). The ADC values was measured on Gd+ lesions, on all lesions showing restricted diffusion in DWI, 2 Gd− lesions, and 1 area of normal-appearing white matter (NAWM)in all MRI studies. To capture the most restricted part of the lesions, ADC measurements were performed with a sufficiently large regions-of-interest (ROIs) that included the lesions and the perilesional edema (Fig. 1) and minimum ADC values were used. This was especially useful for lesions demonstrating more heterogeneity. The maximum and average ADC values represent contribution from inflammatory edema in active lesions or gliosis in chronic lesions..
Method used for screening of lesions and measurement of ADC values
in ADC map.
2.3. Statistical analysis
Statistical analysis was performed using GraphPad Prism version 6.00 for Windows (GraphPad Prism version 6.01 for Windows). Given nonparametric pattern of measured ADC values as continuous variables, Mann-Whitney rank sum test and Wilcoxon matched-paired signed rank test were performed to compare the ADC values of lesions and NAWM. Differences in the distribution of categorical variables were also tested for statistical significance using the Fisher's exact tests. Significance was set at a P value of less than 0.05.
Between January 2007 and July 2014, out of 466 screened patients, 72 patients (16 men, 56 women) met the inclusion criteria of having an MRI with Gd as well as DWI and available images in the PACS (It is not routine at our institution to administer Gd with every scan and often is at the discretion of the neuroradiologist). A total of 104 MRI studies were included in the analysis. The median patient age at the time of MRI was 41 years (interquartile range-IQR=18.6). The median duration of the disease at the time of the MRI study was 8.5 years (IQR=12.89 years). Median EDSS was 3 (IQR=2). More detailed patient demographic data are displayed in Table 1.
Demographic findings of studied group.
|Characteristic of patients (n=72)||Mean±SD or Number (%)|
|Age at onset||41±12.17|
|Disease course at the time of MRI (104)|
|EDSS Score at the time of MRI||3±1.8|
A total of 379 ADC measurements (ADC values max/min/avg/SD) were recorded on 275 lesions and 104 areas of NAWM. Out of the 275 lesions analyzed, 68 were Gd+ and 207 were Gd−. In visual assessment of DWI imaging and ADC map, 34 lesions appeared to show diffusion restriction and 241 showed no restriction (Table 2).
Numbers of the different types of studied lesions.
|Data analyzed||Gd+ lesions||Gd− lesions||Total|
|Restricted lesions in ADC map||23||11||34|
|Non-restricted lesions in ADC map||45||196||241|
Diffusion restriction corresponds to lower ADC values (Le Bihan et al., 1986). As MS lesions show heterogeneous diffusion, the minimum ADC values from ROI were used as a quantitative marker for restricted diffusion depicting the most restricted part of each lesion. The median of ADC-minimum values was 626×10−6 mm2/s (IQR=109) for Gd+ and 936×10−6 mm2/s (IQR=213) for Gd− lesions. Corresponding values in the NAWM was 646.5×10−6 mm2/s (IQR=60). Pairing of data were performed between lesional and NAWM measurements in each MRI. There was a significant difference between median of ADC-minimum of Gd+ lesions versus NAWM (Difference =−20.50×10−6 mm2/s, P=0.0209 in unpaired and P=0.0126 in paired test). The median of ADC-minimum values in the DWI restricted and non-restricted group were 600×10−6 mm2/s (IQR=124.8) and 830×10−6 mm2/s (IQR=254) respectively (Difference=230×10−6 mm2/s, P=<0.0001). There was also a significant difference between median of ADC-minimum values of lesions showing restriction versus NAWM (Difference=−46×10−6 mm2/s, P=0.0093). (Table 3, Fig. 2).
Analysis of ADC metrics of different types of lesions.
|Median 1/IQR (×10−6 mm2/s)||Median2/IQR (×10−6 mm2/s)||Actual Diff (Mann-Whitney)/Median of Diff (Wilcoxon test)||P value (Mann- Whitney)/Median of Diff (Wilcoxon test)||Sig. P<0.05|
|Gd+ vs. NAWM||ADC min||626/109||646.5/60||−20.50/−20.50||0.0209/0.0126||Yes|
|Gd− vs. NAWM||ADC min||936/213||646.5/60||289.5/268.5||<0.0001/<0.0001||Yes|
|Gd+ vs. Gd−||ADC min||626/109||936/213||−310/−291.5||<0.0001/<0.0001||Yes|
ADC-minimum of Gd enhancing lesions vs. NAWM and non-enhancing lesions.
Analysis of the association of observed numbers of restricted lesions in ADC map with number of Gd+ lesions was done with Fisher’s test. Most restricted lesions showed enhancement in CE T1-weighted images (high specificity≥94%), but many Gd+ lesions showed no evidence of restricted diffusion in ADC-map (low sensitivity≤34%) (Table 4, Fig. 3). So, a visually evident lesion with restricted diffusion (hyperintense in DWI and hypointense in the ADC map) strongly predicted Gd+ lesion. However, though many Gd+ lesions quantitatively showed reduced ADC values, these changes were difficult to appreciate visually. This resulted in missing many Gd+ lesions if one were to assess activity using only DWI rather than Gd− enhancement..
Sensitivity and specificity of restriction in ADC map for Gd enhancement.
|ADC Restricted lesions||23||11|
|ADC non-Restricted lesions||≥45||≥196|
|Sensitivity and specificity||Fisher’s exact test (Significant P<0.05)|
|Positive predictive value||0.6765|
|Negative predictive value||≥0.8133|
Visually apparent ADC Restricted lesions in DWI and enhancement pattern.
We also observed that the median of maximum and average of ADC values were significantly higher in Gd+ vs. Gd− lesions compared to the NAWM. The mean ADC max/avg was 1293(IQR=439)/897(IQR=148)×10−6 mm2/s for Gd+ lesions, 1167(IQR=315)/1075(IQR=257)×10−6 mm2/s for Gd− lesions and 800(IQR=130)/708.5(IQR=86)×10−6 mm2/s for corresponding NAWM. The median of SD of Gd+ lesions was also significantly higher than Gd− lesions illustrating higher values in maximum ADC and lower values in minimum ADC in Gd+ vs. Gd− lesions (Table 3, Fig. 4).
ADC-max, ADC-average with Standard Deviation of Gad enhancing vs. non-enhancing lesions and NAWM.
A total of 43 relapses were recorded during the study period. In 29 recorded relapses the enhancing or restricted lesions were not able to explain anatomically the presenting symptoms or signs at the time of relapse; i.e. they were considered “asymptomatic”. In the other 12 relapses enhancing lesions seen at the time of MRI were deemed related to the presenting symptoms or signs and thus were considered “symptomatic”. Out of those, 7 relapses correlated with lesions that showed both restricted diffusion on the DWI as well as Gd− enhancement and the other 5 lesions were non-restricted in DWI. There were 2 relapses solely related to the lesions showing restricted diffusion without evidence of corresponding Gd enhancement (Fig. 5). In addition, there was a weakly significant difference (P=0.042) between median ADC min of symptomatic vs. asymptomatic lesions in the Gd+ lesions. Our data did not show a significant difference between ADC maximum/average of symptomatic versus asymptomatic lesions (Table 5, Fig. 6) in the Gd+ lesions...
Symptomatic vs asymptomatic lesions in Gd enhancing or restricted lesions.
Analysis of ADC values of symptomatic vs. asymptomatic lesions in enhancing lesions.
|Median 1/IQR Symptomatic (×10−6 mm2/s)||Median 2/IQR Asymptomatic (×10−6 mm2/s)||Actual Diff (Mann-Whitney)||P value (Mann- Whitney)||Sig. P<0.05|
|Gd+ symptomatic vs. Gd+ asymptomatic lesions||ADC min||578.5/87.5||630/100||−51||0.0421||Yes|
ADC values of symptomatic vs asymptomatic enhancing lesions.
DWI as an in vivo measure of molecular motion of water (Le Bihan et al., 1986), has been widely used to diagnose acute ischemic infarction (Warach et al., 1996) and also to detect diffusion alterations in active inflammatory lesions (Filippi et al., 2003). DWI can provide quantitative estimates of tissue damage in MS because the permeability and structural barriers to the diffusion of water are changed in the disease. DWI and analysis of the apparent diffusion coefficient (ADC) have provided evidence for subtle progressive alterations in tissue integrity several weeks before focal leakage of the BBB and plaque formation (Werring et al., 2000). Previous limited studies have shown a possible correlation between lesion enhancement and their degree of restricted diffusion (Eisele et al., 2012). Our data is in keeping with other studies that have shown that diffusivity is higher in chronic MS plaques than in NAWM, and diffusion characteristics of chronic lesions differ from those of acute lesions (Larsson et al, 1992 and Christiansen et al, 1993).
In prior limited studies of patients with MS and acute disseminated encephalomyelitis (ADEM), a reduction of the ADC has been documented in the very early phase in acute MS lesions, with cytotoxic edema mimicking the radiological features of acute stroke (Balashov et al, 2011, Bugnicourt et al, 2010, and Rosso et al, 2006). The possible mechanisms of reduced ADC include infiltration of inflammatory cells (mainly T-lymphocytes and macrophages/microglial cells) and associated macromolecules as well as cytotoxic cell swelling and disturbances of energy metabolism, , leading to reduced extracellular space (Eisele et al, 2012, Balashov et al, 2011, Rosso et al, 2006, Lucchinetti et al, 2000, Rigby et al, 2012, and Tievsky et al, 1999).
The reduced ADC signal is transient and appears only in the early stage of an acute lesion and may revert to normal or increased signal within one to two weeks (Eisele et al, 2012, Balashov et al, 2011, and Rigby et al, 2012). We believe the ADC restriction might appear before the stage of enhancement and normalizes much before the disappearance of enhancement. In our study 2 lesions showing restricted diffusion without enhancement might elude to this concept. ADC restriction is present before MRI signs of tissue destruction become prominent and might be predominantly related to parenchymal inflammation, in keeping with mitochondrial dysfunction and subsequent electrical compromise in ADC restricted lesions. Theoretically, these events might even take place in the absence of demyelination, which may be more prominent in slightly later stages of the lesion development (Eisele et al., 2012). In a serial MRI study, MS patients presenting with new symptoms and an associated lesion with reduced ADC, a characteristic sequence of signal-intensity changes was observed: 1) days 0–7: slight T2 hyperintensity and prominent ADC restriction (maximum, −66%), faint or no enhancement on postcontrast T1-weighted images; 2) days 7–10: prominent T2 hyperintensity and contrast enhancement, ADC normalization/pseudonormalization; 3) up to 4 weeks: elevated ADC values, prominent enhancement on post-contrast images; 4) after 4 weeks: partial reversibility of T2 hyperintensity, ADC elevation, and resolution of contrast enhancement (Eisele et al., 2012).
One important observation from our study was the difficulty of appreciating diffusion restriction on visual analysis of the images, though the quantitative minimum ADC values correlated well with Gd enhancement. Most lesions showing restriction in ADC maps upon visual analysis also showed Gd enhancement (specificity>94%), but relying only on the visual appearance of lesions showing restricted diffusion was a weak screening method to find Gd+ lesions, as many Gd+ lesions did not correlate with restricted diffusion on the ADC map (sensitivity<34%).
Our data did show that Gd+ lesions had lower ADC values than Gd− lesions that might differentiate active “symptomatic” lesions from older inactive or “asymptomatic” lesions. The median ADC-minimum value of Gd+ lesions was significantly, though slightly, lower than NAWM. This slight difference might not be easily visible in visual screening of DWI/ADC map, but this difference is more striking when comparing Gd+ and Gd− lesions, being lower in the former.
We observed that the ADC max of Gd+ lesions is significantly higher than ADC max of NAWM and even somewhat higher than in Gd− lesions making it a good screening tool for finding demyelinating lesions, but the difference with Gd− lesions is weak, making it a poor differentiation method between Gd+ and Gd− lesions. However, we noted wider differences between ADC values of Gd+ vs. Gd− lesions. This discrepancy confirms the variability of tissue damage in new enhancing lesions and might reflect different lesion ages (Dousset et al., 1998). A recent cohort of CIS patients suggested hyperintensity on DWI as a possible screening method for enhancing lesions (Lo et al., 2014), but most of the DWI hyperintense lesions of MS patients are old and inactive. Some areas of active lesions have elevated diffusion resulting from disruption of myelin and increased extracellular space (vasogenic) edema (Roychowdhury et al, 2000, Tievsky et al, 1999, Lo et al, 2014, and Castriota-Scanderbeg et al, 2002). The hyperintensity due to alterations of water diffusion in the lesions is more sensitive and lasts longer (may persist several months) than lesion enhancement due to transient blood-brain-barrier (BBB) disruption (which usually lasts 4–6 weeks) (Roychowdhury et al, 2000, Eisele et al, 2012, and Rosso et al, 2006). However, the persistent hyperintensity on DWI may suggest residual extracellular edema with increased diffusion, prolongation of T2 relaxation time, and the T2 shine-through effect (Lo et al., 2014).
In our study, most Gd+ lesions or visually restricted lesions did not correlate with the clinical presentation (symptoms or signs) of patients, but we observed lower ADC minimum of symptomatic Gd+ vs. asymptomatic lesions. Our study illustrates again the clinical-imaging paradox in conventional MRI sequences (with many enhancing or apparent restricted lesions being asymptomatic) and might suggest the added value of DWI quantitative data in structural study of active demyelinating lesions.
One of the main limitations of this study was that it was a retrospective study; the group was limited to what we encounter in real clinical setting, including patients of different clinical stages (i.e. early, relapsing or progressive MS) on various previous treatments including corticosteroids and immunomodulators. Another is that Gd is not routinely used for either diagnosing MS or monitoring the response to therapy at our institution. There was also a wide range between the timing of the MRI study and the presenting clinical symptoms in assessing “symptomatic” vs. “asymptomatic” lesions. Further prospective studies in a more homogenous group and more closely timed scans to clinical relapses would be needed to validate our results. Although an in-house standard MRI protocol was being followed, different MRI machines and technicians performed the studies, which produced variability in the picture quality. The definition of the ROI in many Gd+ lesions varied and contained NAWM in perilesional areas. This affected measurements, especially min & avg.-ADC of lesions. We only measured 2 Gd− lesions for every patient as a sample of Gd− lesions and more sampling might have yielded different results since the correlation of restriction in ADC-map and Gd+ lesions with clinical presentation was calculated in this limited group of measurements. Out of 104 MRI studies, 45 studies were done while patients were on DMDs. This study was underpowered to assess possible effect of DMDs on the DWI. DWI is not routinely used to evaluate the spinal cord in conventional MRI studies and was not included in our study given the technical difficulty and poor resolution of images. Spinal cord lesions tend to be more likely symptomatic and differences in diffusivity of Gd+ lesions may correlate better than they did in brain lesions.
Our study shows that diffusivity is heterogeneous in Gd+ MS lesions, which have lower minimum ADC values (restricted) and higher maximum ADC values compared to Gd− lesions as well as NAWM. Lesions that visually restrict diffusion on DWI (hyperintense in DWI and hypointense in the ADC map) often enhance with Gd on T1-weighted images (high specificity). However, though most Gd+ lesions quantitatively show decreased ADC values, it is difficult to visually appreciate diffusion restriction in these lesions resulting in a low sensitivity to predict the presence of enhancement. This makes the DWI/ADC a poor surrogate marker for enhancement and practically a Gd− enhanced study cannot be replaced by DWI/ADC. Lower ADC values of lesions did correlate more tightly with the lesions that were symptomatic. Further prospective studies in a more homogenous group and more closely timed scans to clinical relapses would be needed to validate this correlation.
Conflict of interest
We certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.
- Balashov et al., 2011 K.E. Balashov, L.L. Aung, S. Dhib-Jalbut, et al. Acute multiple sclerosis lesion: conversion of restricted diffusion due to vasogenic edema. J. Neuroimaging. 2011;21:202-204 Crossref
- Bugnicourt et al., 2010 J.-M. Bugnicourt, P.-Y. Garcia, P. Monet, et al. Teaching neuroimages: marked reduced apparent diffusion coefficient in acute multiple sclerosis lesion. Neurology. 2010;74:e87 Crossref
- Castriota-Scanderbeg et al., 2002 A. Castriota-Scanderbeg, U. Sabatini, F. Fasano, et al. Diffusion of water in large demyelinating lesions: a follow-up study. Neuroradiology. 2002;44:764-767 Crossref
- Christiansen et al., 1993 P. Christiansen, P. Gideon, C. Thomsen, et al. Increased water self-diffusion in chronic plaques and in apparently normal white matter in patients with multiple sclerosis. Acta Neurol. Scand.. 1993;87:195-199
- Dousset et al., 1998 V. Dousset, A. Gayou, B. Brochet, et al. Early structural changes in acute MS lesions assessed by serial magnetization transfer studies. Neurology. 1998;51:1150-1155 Crossref
- Eisele et al., 2012 P. Eisele, K. Szabo, M. Griebe, et al. Reduced diffusion in a subset of acute MS lesions: a serial Multiparametric MRI Study. Am. J. Neuroradiol.. 2012;33:1369-1373 Crossref
- Filippi et al., 2000 M. Filippi, G. Iannucci, M. Cercignani, et al. A quantitative study of water diffusion in multiple sclerosis lesions and normal-appearing white matter using echo-planar imaging. Arch. Neurol.. 2000;57:1017-1021 Crossref
- Filippi et al., 2003 M. Filippi, M.A. Rocca, G. Comi. The use of quantitative magnetic-resonance-based techniques to monitor the evolution of multiple sclerosis. Lancet Neurol.. 2003;2:337-346 Crossref
- Freedman et al., 2013 M. Freedman, D. Selchen, D. Arnold, et al. Treatment optimization in MS: canadian MS working group updated recommendations. Can. J. Neurol. Sci.. 2013;40:307-323 Crossref
- , GraphPad Prism version 6.01 for Windows, GraphPad Softwarewww.graphpad.com. La Jolla California USA,: GraphPad software Inc.
- iMed, 2014 iMed, 2014. - About iMed 〈http://www.imed.org/en/about_imed/about_imed.html〉 (accessed 18.11.14).
- Kanal et al., 1990 E. Kanal, F.G. Shellock, L. Talagala. Safety considerations in MR imaging. Radiology. 1990;176:593-606 Crossref
- Kurtzke, 1983 J.F. Kurtzke. Rating neurologic impairment in multiple sclerosis an expanded disability status scale (EDSS). Neurology. 1983;33 (1444–1444)
- Larsson et al., 1992 H.B.W. Larsson, C. Thomsen, J. Frederiksen, et al. In vivo magnetic resonance diffusion measurement in the brain of patients with multiple sclerosis. Magn. Reson. Imaging. 1992;10:7-12 Crossref
- Le Bihan et al., 1986 D. Le Bihan, E. Breton, D. Lallemand, et al. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology. 1986;161:401-407 Crossref
- Lo et al., 2014 C.-P. Lo, H.-W. Kao, S.-Y. Chen, et al. Comparison of diffusion-weighted imaging and contrast-enhanced T1-weighted imaging on a single baseline MRI for demonstrating dissemination in time in multiple sclerosis. BMC Neurol.. 2014;14:100 Crossref
- Lublin et al., 2014 F.D. Lublin, S.C. Reingold, J.A. Cohen, et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology. 2014;83:278-286 Crossref
- Lucchinetti et al., 2000 C. Lucchinetti, W. Brück, J. Parisi, et al. Heterogeneity of multiple sclerosis lesions: implications for the pathogenesis of demyelination. Ann. Neurol.. 2000;47:707-717 Crossref
- Manual on Contrast Media v9, 2014 Manual on Contrast Media v9, 2014. - American College of Radiology 〈http://www.acr.org/quality-safety/resources/contrast-manual〉 (accessed 17/11/14).
- McDonald et al., 2015 R.J. McDonald, J.S. McDonald, D.F. Kallmes, et al. Intracranial gadolinium deposition after contrast-enhanced MR imaging. Radiology. 2015;275:772-782 Crossref
- Miller et al., 1988a D.H. Miller, P. Rudge, G. Johnson, et al. Serial gadolinium enhanced magnetic resonance imaging in multiple sclerosis. Brain. 1988;111:927-939 Crossref
- Miller et al., 1998b D.H. Miller, R.I. Grossman, S.C. Reingold, et al. The role of magnetic resonance techniques in understanding and managing multiple sclerosis. Brain. 1998;121:3-24 Crossref
- Molyneux et al., 1998 P.D. Molyneux, M. Filippi, F. Barkhof, et al. Correlations between monthly enhanced MRI lesion rate and changes in T2 lesion volume in multiple sclerosis. Ann. Neurol.. 1998;43:332-339 Crossref
- MRI Private Costs - Scans & Rates, 2014 MRI Private Costs - Scans & Rates, 2014. Canadian Magnetic Imaging (CMI) 〈http://www.canmagnetic.com/scans-rates/〉 (accessed 17.11.14).
- Polman et al., 2011 C.H. Polman, S.C. Reingold, B. Banwell, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann. Neurol.. 2011;69:292-302 Crossref
- Radiology Information System with Integrated Solutions, 2015 Radiology Information System with Integrated Solutions, 2015. | McKesson 〈http://www.mckesson.com/providers/health-systems/department-solutions/radiology/mckesson-radiology/〉 (accessed 09.09.15).
- Rigby et al., 2012 H. Rigby, W. Maloney, V. Bhan. Neuroimaging highlight - diagnostic considerations in acute ms lesions with restricted diffusion on MRI. Can. J. Neurol. Sci.. 2012;39:525-526 Crossref
- Rosso et al., 2006 C. Rosso, P. Remy, A. Creange, et al. Diffusion-weighted MR imaging characteristics of an acute strokelike form of multiple sclerosis. Am. J. Neuroradiol.. 2006;27:1006-1008
- Roychowdhury et al., 2000 S. Roychowdhury, J.A. Maldjian, R.I. Grossman. Multiple sclerosis: comparison of trace apparent diffusion coefficients with MR enhancement pattern of lesions. Am. J. Neuroradiol.. 2000;21:869-874
- Sormani et al., 2013 M.P. Sormani, J. Rio, M. Tintorè, et al. Scoring treatment response in patients with relapsing multiple sclerosis. Mult. Scler. J.. 2013;19:605-612 Crossref
- Thomsen et al., 2013 H.S. Thomsen, S.K. Morcos, T. Almén, et al. Nephrogenic systemic fibrosis and gadolinium-based contrast media: updated ESUR Contrast Medium Safety Committee guidelines. Eur. Radio.. 2013;23:307-318 Crossref
- Tievsky et al., 1999 A.L. Tievsky, T. Ptak, J. Farkas. Investigation of apparent diffusion coefficient and diffusion tensor anisotropy in acute and chronic multiple sclerosis lesions. Am. J. Neuroradiol.. 1999;20:1491-1499
- Warach et al., 1996 S. Warach, J.F. Dashe, R.R. Edelman. Clinical outcome in ischemic stroke predicted by early diffusion-weighted and perfusion magnetic resonance imaging: a preliminary analysis. J. Cereb. Blood Flow. Metab. J. Int. Soc. Cereb. Blood Flow. Metab.. 1996;16:53-59 Crossref
- Werring et al., 1999 D.J. Werring, C.A. Clark, G.J. Barker, et al. Diffusion tensor imaging of lesions and normal-appearing white matter in multiple sclerosis. Neurology. 1999;52:1626-1632
- Werring et al., 2000 D.J. Werring, D. Brassat, A.G. Droogan, et al. The pathogenesis of lesions and normal-appearing white matter changes in multiple sclerosis. Brain. 2000;123:1667-1676 Crossref
a University of Ottawa, Canada
b Department of Radiology, University of Ottawa, Department of Medical Imaging, The Ottawa Hospital, Canada
c Ottawa Hospital, General Campus, University of Ottawa, Canada
d The Ottawa Hospital Research Institute and the University of Ottawa, Canada
⁎ Corresponding author.
© 2016 Elsevier B.V., All rights reserved.