Abstract
Background: One of the diagnostic imaging hallmarks of MS is the uptake of IV administered contrast material in new lesions in the brain, signaling blood–brain barrier breakdown and active inflammation. Many clinical drug trials are designed based on the assumption that lesion enhancement on MRI remains visible on average for 1 month. For practical reasons, few serial MRI studies of patients with MS have been performed at intervals shorter than 4 weeks.
Methods: The authors performed a year-long longitudinal study in 26 patients with relapsing-remitting MS (RRMS), which comprised an initial phase of MRI follow-up at weekly intervals for 8 weeks, followed by imaging every other week for another 16 weeks, and monthly thereafter. They present a quantitative analysis (using a supervised interactive thresholding procedure) of new enhancing lesions appearing during the first 6 weeks in this cohort and evaluated from the time of first detection until enhancement was no longer seen.
Results: The average duration of Gd-DTPA enhancement in individual new lesions was 3.07 weeks (median, 2 weeks). Significant correlations were demonstrated between the duration of contrast enhancement or initial growth rates and lesion volumes. Different lesions in the same patient appeared to develop largely independent of each other and demonstrated a large range in the duration of enhancement during the acute phase of their evolution.
Conclusions: The average duration of blood–brain barrier impairment in RRMS is shorter than earlier estimates. Early lesion growth parameters may predict final lesion size. Within-patient heterogeneity of lesion evolution suggests that individual lesions develop independently.
One of the diagnostic imaging hallmarks of MS is the uptake of IV administered contrast material in new lesions of the brain, signaling blood–brain barrier (BBB) breakdown. This feature, observed on MRI, has been exploited in the context of clinical drug trials for this disease. Many of these trials are designed on the assumption that lesion enhancement remains visible on average for 1 month.1-4⇓⇓⇓ On monthly scans, 60 to 86% of all enhancing lesions are seen only on a single examination.3,5-12⇓⇓⇓⇓⇓⇓⇓⇓ Animal studies of experimental allergic encephalomyelitis (EAE), a condition not identical to MS but often used as a model of this disease, have demonstrated that enhancement consistently lasted less than 5 days in acute EAE whereas many lesions enhanced for less than 15 days in a chronic EAE model.13,14⇓
For practical reasons, few serial MRI studies of patients with MS have been performed at intervals shorter than 4 weeks. Aside from the patients described in this article, only two studies reported patients followed at weekly intervals: one study presented results from three patients followed with weekly MRI for 3 months; the other described six patients imaged weekly for 2 months.15,16⇓
We performed a year-long longitudinal study in 26 patients with relapsing-remitting MS (RRMS), which comprised an initial phase of MRI follow-up at weekly intervals for 8 weeks, followed by imaging every other week for another 16 weeks, and monthly thereafter.
We present a quantitative analysis of each new enhancing lesion (n = 113) appearing during the first 6 weeks in this cohort and followed until enhancement was no longer seen. Three findings in this work have major implications for understanding the natural history of RRMS and the design of clinical drug trials:
-
The average duration of enhancement of MS lesions was lower than estimates currently used and closer to what is observed in animal models of this disease.
-
Significant correlations were demonstrated between the duration of contrast enhancement and lesion volumes or growth rates.
-
Different lesions in a same patient appeared to develop largely independent of each other and demonstrated large variation in the duration of enhancement during the acute phase of their evolution.
Methods.
Study population.
Twenty-six patients with RRMS were followed over a 1-year period by MRI at weekly intervals for 8 weeks, followed by MRI examinations every other week for another 16 weeks, and at monthly intervals for the rest of the year.4,17-19⇓⇓⇓ The data for this study were acquired between 1990 and 1992. All patients gave informed consent to participate in the study by signing a consent form approved by our institutional review board.
Clinical examinations were performed monthly and at the time of clinical exacerbations. Patients previously treated with immunosuppressive, cytotoxic, or immunomodulatory drugs were excluded. At the time of the study interferons were not yet indicated for the treatment of MS and were therefore not given to any of the patients during this study. Patients may have received corticosteroids in the past, but not within 2 months of admission into the study. Patients with major illnesses other than MS were excluded. In addition, patients with claustrophobia as well as patients with pacemakers, aneurysm clips, other metal implants, or shrapnel fragments were excluded. Table 1 summarizes the clinical profile of the patients. Nine of the 26 patients in this study received a total of 34 short courses of prednisone or IV methylprednisone treatment for acute attacks during the course of this study. A total of 113 new enhancing lesions were detected during the initial 6-week follow-up period in 13 of the 26 patients. Ninety-one lesions were detected in nine patients who did not receive corticotherapy during the study; one new lesion was detected in a patient who received corticotherapy only after the lesion stopped enhancing and after the initial 6 weeks of follow-up; the remaining 21 lesions were potentially affected by corticotherapy as they appeared during or following corticotherapy, or had not stopped enhancing before corticotherapy was administered in the remaining four patients. For the purpose of defining the natural history of enhancing lesions in RRMS, we summarize the results from the 92 new enhancing lesions, which were unaffected by corticotherapy during the study, and were detected in 10 patients with RRMS.
Table 1 Patient demographics and disease characteristics
MRI examinations.
Each MRI examination included sagittal T1-weighted localizer images, dual-echo (proton density [PD]–weighted and T2-weighted) axial images, and contrast-enhanced T1-weighted axial images, which were acquired on a 1.5 Tesla MRI system (Signa, General Electric Medical Systems, Milwaukee, WI) and transferred to a workstation (Sun Microsystems, Mountain View, CA) for further processing. This report focuses on the evaluation of postcontrast T1-weighted images. The precontrast T1-weighted, PD-weighted, and T2-weighted images were used only to help confirm the presence of new enhancing lesions.
Postcontrast transaxial T1-weighted images were acquired without delay upon administration of an IV bolus of 10 mL of 500 mmol/L of gadopentate dimeglumine (Gd-DTPA) (Magnevist, Berlex Laboratories, Cedar Knolls, NJ). Using a conventional spin-echo sequence, images were acquired with one excitation, a repetition time (TR) of 600 msec, and an echo time (TE) of 19 msec. Slice thickness was 4 mm with a gap of 1 mm between slices. Each slice covered a 24 cm × 24 cm field of view and was acquired using a 256 × 192 matrix. Images of each slice were reconstructed at 256 × 256 resolution after zero filling of the acquisition matrix resulting in a nominal voxel size of 0.94 × 0.94 × 5 mm3 (including gap between slices); i.e., 4.4 mm3/voxel. The phase offset multiplanar option enabled the acquisition of images from 24 brain sections within approximately 2.5 minutes.
The sagittal precontrast T1-weighted images were acquired with identical parameters. Axial dual-echo images were acquired from contiguous 3-mm-thick sections, using two interleaved conventional spin-echo sequences.20
Image analysis.
Enhancing lesions were defined by a neuroradiologist (F.C.) on contrast-enhanced images as contiguous regions showing hyperintensity relative to the surrounding white matter. Hyperintensity can also occur on nonenhanced T1-weighted images in some MS lesions.21-23⇓⇓ To avoid misclassification with this subgroup of lesions, the precontrast T1-weighted images were compared to those obtained after Gd-DTPA injection.
Only signal abnormalities larger or equal to two voxels were considered lesions.
New enhancing lesions were defined as enhancing lesions not detected in a patient’s first MRI study. We described only the course of new enhancing lesions appearing during the first 6 weeks in order to assure weekly follow-up immediately after lesion detection.
The lesion time point (LTP) was defined as the snapshot of a lesion at one time point.
New enhancing lesions were subdivided into two groups: nodular enhancing lesions (NEL) and ring enhancing lesions (REL). NEL demonstrated relatively homogeneous enhancement over the whole lesion at all LTP. REL showed a central, nonenhancing portion with a peripheral enhancing rim at one or more LTP. Any enhanced arc of circle of more than 180° was considered a REL. Our calculations only include enhancing pixels, and therefore in REL the hypo-/isointense central pixels are not included.
Duration of enhancement was defined for each lesion as the time elapsed between the scan prior to the lesion’s initial detection and the time when the last enhancement was noted. By this definition, the duration of enhancement of a lesion visible only on one scan was 1 week. In truth, the persistence of enhancement in a lesion visible only on one of the MRI examinations performed at weekly intervals could be anywhere between 1 and 13 days.
Time integrated volume of enhancement (TIVE) was computed as the sum of enhancing pixels over the duration of enhancement for an individual new enhancing lesion. Measured or estimated (see below) weekly LTP were used for this computation. For REL we also calculated the total time integrated volume (TTIV) by adding the nonenhancing pixels of a lesion at each LTP to the TIVE.
Average enhanced size per LTP was defined for each new enhancing lesion as TIVE divided by the duration of enhancement (in weeks).
Lesion volumes were estimated using a supervised, interactive thresholding procedure. A local (lesion by lesion) thresholding technique leveraged the combined advantages of expert lesion identification by a radiologist and computer-aided delineation of lesion boundaries. The new lesions were segmented (slice by slice) at each time point (LTP) using the MRX software package, a joint project of General Electric Medical Systems (Milwaukee, WI), and the Surgical Planning Laboratory of the Brigham and Women’s Hospital. The MRX software includes an interactive image browser with drawing, erasing, zooming, and quantification tools. The threshold for the segmentation of each enhanced lesion was determined with respect to the signal intensity of the adjacent normal appearing white matter. Enhancing pixels were grouped into individual lesions using a four-neighbor connectivity algorithm.24 The three-dimensional lesion volume at a single time point was then computed by summing contiguous voxels labeled as lesion over all slices.
A single observer (F.C.) performed the detection of the optimum threshold for each LTP. To estimate the influence of this operator-dependent step on measurement error, 40 randomly chosen lesions (20 NEL and 20 REL) were independently evaluated by two observers, one of whom repeated the evaluation approximately 3 weeks later (F.C.). The agreement index between pairs of measurements was calculated as previously described.25 The median interobserver agreement index was 80% whereas the median intraobserver agreement index was 87%.
Statistical analysis.
Descriptive statistics for the patient sample are shown in table 1. For a given patient, if the LTP were not uniformly spaced at 7- or 14-day intervals, the lesion loads corresponding to weekly intervals were estimated by linear interpolation. For instance, if images were taken at days 0, 8, and 14, the volume for the second time point at 1 week (day 7) was estimated using linear interpolation between the volume at day 0 and the volume at day 8. New enhancing lesions detected during the first 6 weeks of follow-up were tracked until enhancement was no longer seen. Volumes of 17 LTP within the weekly monitoring period were inferred using linear interpolation.
Twenty-seven LTP were collected 2 weeks after the previous MRI. Again, the volumes at weekly LTP were estimated by linear interpolation. In summary, a total of 44 of 319 lesion volumes were estimated from actual measurements by linear interpolation to regularize the sampling of our data in the time domain. The average sampling error of these 44 LTP was 5.43 days (SD 4.4 days, range 2 to 14 days). Correlation coefficients (ρx,y= cov(X,Y)/ς x.ς y) and linear fitting parameters were calculated to assess the relationship between enhancement duration and lesion volume.
Paired and unpaired one-tailed Student’s t-tests were used to compare volume distributions of different subgroups.
Results.
During the first 6 weeks of monitoring, 113 new enhancing lesions were detected in 13 of the 26 patients with RRMS. Overall, the 113 lesions were followed for a total of 319 time points (LTP). Excluding the lesions potentially affected by corticotherapy, 92 new enhancing lesions were detected in 10 of the 23 patients and were followed for a total of 248 LTP (“natural history” group). Each lesion was evaluated from the time of first detection until enhancement was no longer seen. The results are reported after regularizing the data with linear interpolation. Unless otherwise noted, the following results refer only to the 92 lesions in the natural history group that were detected in 10 patients with RRMS.
Duration of enhancement.
The average duration of enhancement of new lesions was 3.07 weeks (median 2 weeks); excluding a pseudotumoral lesion, the average duration of new lesions was 2.96 weeks (median 2 weeks).
The distribution of the duration of enhancement of new enhancing lesions is shown in figure 1. Twenty-six new enhancing lesions (28.26%) were visible only on one MR examination; 25 new enhancing lesions (27.17%) were visible on two consecutive weekly examinations. Hence, 55.43% of new enhancing lesions demonstrated enhancement for less than 3 weeks when followed with MRI at weekly intervals. Twelve new enhancing lesions showed contrast enhancement on three weekly examinations and 10 were visible on four sequential weekly MRI examinations. Only 3 (3.26%) of the lesions enhanced longer than 2 months. The longest period of enhancement was 13 weeks and concerned a ring-shaped pseudotumoral lesion (another such lesion was present in the corticosteroid group). Anecdotally, this pseudotumoral REL was visible on PD-weighted and T2-weighted images 7 days before enhancement.
Figure 1. Distribution of new enhancing lesions in patients with relapsing-remitting MS (RRMS) according to enhancement duration. The distribution of enhancement duration in new lesions followed at weekly intervals is non-gaussian. The distribution of new enhancing lesions according to the duration of their enhancement is clearly skewed toward enhancement duration of 2 weeks and less. The salient finding of our study is that over 50% of identifiable new lesions in RRMS have a phase of early enhancement visible in only one or two weekly scans. Dark gray: subgroup of lesions potentially affected by corticotherapy (n = 21); light gray: natural history lesions (n = 92).
Volume of enhancement.
On average, new enhancing lesions reached their maximum enhancement size in 1.74 weeks (median 1 week) and disappeared gradually thereafter in 2.33 weeks (median 2 weeks). The duration of the growth phase was shorter than that of the waning phase (p < 0.005, paired one-tailed Student’s t-test). We estimated the overall volume of enhancement within a lesion by integrating the volume of enhanced tissue over the duration of enhancement within each lesion. The average TIVE for the 92 new enhancing lesions (including the large pseudotumoral lesion) was 0.2 mL (74.93 voxels) (median 0.06 mL or 21.5 voxels).
Correlation between duration of enhancement and enhancement size.
A significant correlation was found between the duration of enhancement of individual new lesions and their maximal enhancement size (r = 0.81) (figure 2, A and B). When applying an exponential function to model the relationship between these two parameters a correlation coefficient of r = 0.92 was obtained.
Figure 2. Correlation between size and duration of enhancement. The persistence of enhancement is related to maximum enhanced volume (A, B) and to the initial growth in enhancement volume over the first week of observation (C). The maximum size of enhancement of each new lesion during follow-up was averaged for groups of lesions with similar duration of enhancement; averages and standard deviations are plotted on a logarithmic scale. (A) A significant correlation was found between the duration of enhancement of individual new lesions and their maximum size of enhancement (r = 0.81). When the data were fit with an exponential function (thick line in figure) the correlation coefficient was 0.92. (B) The distribution of nodular enhancing lesions (NEL [squares]) and ring enhancing lesions (REL [triangles]) with respect to duration and maximum size of enhancement. (C) The relationship between the initial growth of a lesion (between time 1 and time 2) and the duration of enhancement for NEL and REL.
The duration of enhancement was correlated with the average size of enhancement per LTP (r = 0.74), the growth of enhancement volume within the first week after detection (r = 0.66) (figure 2C), and the growth rate of enhancement volume from beginning of enhancement until maximum enhancement size (r = 0.91).
Ring enhancing lesions vs nodular enhancing lesions.
Of the 92 new enhancing lesions, 69 (75%) were NEL; i.e., had nodular appearance from beginning to end. Twenty-three lesions (25%) demonstrated ring enhancement at some point in their evolution, and were therefore classified as REL: 11 out of the 23 showed nodular enhancement at first detection, but later developed ring enhancement, and reverted to a nodular pattern by the end of their enhancing phase; 5 lesions demonstrated ring enhancement throughout their evolution; 3 lesions showed ring enhancement at first detection but demonstrated a nodular pattern at the end of their enhancing phase; 4 lesions demonstrated nodular enhancement early in their evolution and were ring-shaped by the end of their enhancing phase.
REL were larger and had longer persistence of enhancement than NEL (p < 0.005, unpaired Student’s t-test assuming unequal variances) (see figure 2B).
Differences in volumetric measurement between REL and NEL are shown in table 2.
Table 2 Lesion characteristics
All the new lesions with a growth rate of enhancement of more than 15 voxels/week to reach their maximum size of enhancement (n = 9) were ring-shaped lesions. All the new lesions with a maximum size of enhancement more than 50 voxels (n = 6) were ring-shaped lesions. All the new lesions with a time integrated volume more than 200 voxels (n = 5) were ring shaped. All the new lesions with duration of enhancement more than 7 weeks (n = 4) were ring-shaped lesions. The smallest duration of enhancement for a REL was 1 week and concerned one new lesion. The longest duration of enhancement for a nodular lesion was 7 weeks and concerned two new lesions. These two lesions presented the largest time integrated volume of nodular lesions (151 and 191 voxels).
One patient presented a pseudotumoral ring-shaped lesion with a total time integrated volume (with regularization using linear interpolation) of 4,961 voxels (approximately 21.8 mL) and a TIVE of 2,402 voxels (approximately 11.6 mL).
Heterogeneity in the shape and duration of new lesions.
Seven patients presented more than two new lesions during the initial 6-week follow-up period. Five of these seven patients presented both NEL and REL with a large variability in their duration of enhancement. One of the two remaining patients presented three REL and the other one presented three NEL. Most new lesions appeared to evolve independently of changes in other new lesions. For example, a patient presented four new lesions: three NEL with a TIVE less than 10 voxels and one pseudotumoral REL with a TIVE larger than 1,000 voxels. The volumetric evolution of all new lesions in this patient is shown in figure 3A. The three NEL were visible at different time points and each of these lesions was visible on only one MRI examination. This same patient’s REL, on the other hand, presented enhancement for 13 weeks.
Figure 3. Heterogeneity of enhancement duration and size. The temporal evolution of new enhancing lesions is illustrated in two patients with relapsing-remitting MS followed at weekly intervals during the first 8 weeks and then every other week. Each lesion that appeared during the first 6 weeks was evaluated from the time of first detection until enhancement was no longer seen. These illustrations demonstrate the presence of extensive heterogeneity in duration and size of lesion enhancement found within individual patients during the natural course of MS. (A) Follow-up of all new enhancing lesions that appeared during the first 6 weeks of follow-up in a patient with MS. This chart is obtained from the original lesion time points. A logarithmic scale on the y axis is used to accommodate the large variation in size of different new lesions in this patient. Each symbol in the plot represents an individual lesion. The large new lesion (open square) was a ring-shaped pseudotumoral lesion with a time integrated volume of enhancement more than 1,500 voxels and a duration of enhancement of 13 weeks. The three small new enhancing lesions (filled symbols) were all nodular enhancing lesions. Their average time integrated volume was 5.67 voxels and their enhancement was visible only on one weekly scan. (B) Follow-up of 5 out of 40 new enhancing lesions in a patient with MS. Only few new lesions with a similar enhanced size are represented for clarity of the figure. Duration of enhancement for these lesions varied between 1 and 7 weeks.
The median number of new enhancing lesions per patient in the first 6 weeks for the 10 patients was 3.5. Two patients presented very high new lesion activity: one showed 21 new lesions and the other one showed 40 new lesions in the initial 6 weeks. The duration of enhancement of the new lesions in these two patients was not different from the duration of enhancement of new lesions in the eight other patients with new enhancing lesions (p < 0.005): one patient with nine, one with eight, one with four, two with three, one with two, and two with one new enhancing lesion. Figure 3B presents the heterogeneous appearance and evolution of 5 of the 40 new enhancing lesions in one of the two patients with very high new lesion activity.
Discussion.
Enhancing lesions signify BBB impairment and often reflect active perivascular inflammation in MS26-28⇓⇓ as well as in EAE.13,29⇓
New lesions are seen 5 to 10 times more frequently than clinically eloquent relapses in patients with MS.30,31⇓ The number of gadolinium-enhancing lesions is one of the most commonly used MRI-based outcome measures in treatment trials of MS.
Animal studies have demonstrated an approximately linear relationship between the mean number of lesions and the severity of clinical disability at relapse and, in addition, a direct relationship between the duration of clinical relapse and the duration of enhancement with gadolinium for lesions associated with the relapse.14
The salient finding of our study is that the majority of new lesions in RRMS demonstrate enhancement only for 1 or 2 weeks. The distribution of new enhancing lesions according to the duration of their enhancement is skewed toward enhancement duration of 1 week and less, suggesting that more frequent MR follow-up would unveil average duration of enhancement much closer to that found in animal models of the disease.
In EAE, the duration of Gd-DTPA leakage for individual lesions varied from 5 days to 5 weeks in a chronic EAE model and was always less than 5 days in an acute EAE model.13,14⇓ Based on the findings of our study, which are consistent with the findings in models of EAE in animals, it is likely that most new lesions demonstrate BBB breakdown for less than 1 week.
Consistent with our findings, approximately 70% of new lesions in published longitudinal MRI studies performed at monthly intervals were visible on only one scan.3,5,6,8,10-12⇓⇓⇓⇓⇓⇓ Based on these studies, the average duration of enhancement was estimated at approximately 1 month. Our study demonstrates that much higher temporal sampling is required to correctly estimate the average duration of enhancement in MS lesions. In our study, the average duration of enhancement of new lesion (visibility of the enhancement) was 3.14 weeks and the median was 2 weeks. More than 55% of new lesions were visible only on one or two weekly scans.
The measured volume of an enhancing lesion is dependent upon the time between injection and imaging, the amount of gadolinium injected relative to body weight, and the age of the lesion. The majority of lesions show maximal MR signal enhancement on average 29 minutes after IV gadolinium administration.2 In this study, the administration of Gd-DTPA was consistently done immediately before the MRI acquisition and therefore most likely identified only the lesions with high-permeability breakdown of the BBB (high permeability to Gd-DTPA).14,32,33⇓⇓ Although we cannot exclude that a different delay would demonstrate longer average duration of enhancement (i.e., that our measurement lacks sensitivity to detect smaller BBB leakage), it is clear that monthly sampling of our data would yield results similar to those published in the literature. Further frequent MRI studies with long postinjection delay or higher dose of contrast agent would be needed to conclusively address this question.
In a study with 10 patients with early RRMS, the duration of enhancement in 36 lesions was measured and the authors suggested that the larger lesions tend to enhance longer.6 In our study, all measures reflecting lesion size (maximal enhancement size, average size of enhancement) were well correlated to the duration of lesion enhancement. The association between lesion growth in the first week of observation and duration of enhancement, which is most likely indicative of persistent inflammatory activity, suggests that the extent of lesion growth might be predicted by the initial volumetric progression of the lesion.
In the literature, approximately one third of the enhancing lesions are ring-shaped at any one timepoint of MR observation.4,34-36⇓⇓⇓ The current study found a lower occurrence of ring enhancement when excluding patients receiving corticosteroids during lesion follow-up: 23 (25%) new lesions were REL and 69 (75%) were NEL. An analysis of the 21 new enhancing lesions appearing during or after corticotherapy in four of our study’s patients revealed 14 (66.7%) REL and 7 (33.3%) NEL. Combined, 37 of 113 (32.74%) lesions followed in 14 patients were REL, more consistent with previous findings, and suggesting that the four patients requiring corticotherapy might have had more aggressive lesions. No significant difference in duration of enhancement was detected between treated and untreated lesions. It has been reported that when new lesions are followed serially with MR, their enhancement is ring shaped in the majority of lesions older than 30 days. Small new lesions tend to enhance uniformly, corresponding most likely to focal perivenular lesions, whereas larger, older lesions show peripheral ring-like enhancement, in keeping with postmortem histologic findings of myelin breakdown products and inflammatory cells in the same distributions.4,27,37,38⇓⇓⇓ There is a general consensus that the peripheral enhancing portion of a ring-enhancing lesion represents the active, inflammatory area, whereas the nonenhancing central portion represents an older and potentially demyelinated area or gliotic scar.4,39⇓ The ring-shaped lesions presented a maximum enhanced volume, TIVE, and average volume of enhancement for individual lesions significantly larger than that of NEL. Moreover, REL showed significantly longer duration and growth rate of enhancement than NEL. Indeed, our study showed that the majority of new lesions with enhancement duration longer than 1 month were REL. The majority of the ring-shaped lesions were nodular at some point in their evolution (18 of 23 REL); nevertheless, we observed eight enhancing lesions with ring appearance at the time of their detection—i.e., 1 to 6 days after the beginning of the BBB breakdown. Biopsy-proven inflammatory demyelination of the CNS demonstrated early active demyelination in three patients who showed extensive ring-shaped gadolinium enhancement on T1 Gd-DTPA, consistent with our finding that ring-enhanced lesions demonstrate longer enhancement, signifying more persistent inflammatory activity and potentially more severe degeneration. The association of ring enhancement with more destructive lesions has also been previously suggested by others.12,38,40⇓⇓ It has been speculated that diminished magnetization transfer ratio (MTR) reflects diminished myelin content and that hypointensity on T1-weighted images, including both hypointense nonenhancing lesions and the central, hypointense portions of ring-enhancing lesions, corresponds to demyelination. The authors demonstrated that the central regions of ring-enhancing lesions had a lower MTR than the periphery, suggesting that demyelination in MS lesions occurs centrifugally.38 In another study, 12 REL were evaluated with T1-weighted spin-echo and magnetization transfer MRI and the authors concluded that REL may represent the most destructive lesions.12 Again, it appears that early lesion-specific signs might be potentially predictive of ultimate lesion aggressiveness and growth. In our study, immediately after gadolinium administration, one-third of the LTP showed ring-shaped lesion at our spatial resolution. The ring-shaped lesion is also dependent upon the time between injection and imaging. It has been observed that in the majority of enhancing lesions greater than 5 mm in diameter, enhancement 4 minutes after injection was ring-like, but became homogeneous by 16 to 20 minutes.2 A lesion on gadolinium-enhanced T1-weighted images that first appeared as ring enhancement then filled in during the first 55 minutes after gadolinium administration has been also described.41
Some new lesions became ring-shaped in the course of the observed evolution; some apparent NEL were, in fact, part of ring-like structure on contiguous slices. The ring-shaped lesion is likely to represent a quasi-obligatory spatial and temporal event in relation with the concentric evolution of acute destructive and chronic reparatory processes in a lesion.4,42,43⇓⇓ Naturally, the detectability of ring patterns is also a function of image resolution.
Some patients presented very high new lesion activity, whereas others presented no new lesion activity. In a given patient, BBB disturbance showed marked heterogeneity between new lesions. In a given patient, some new lesions could be nodular while others were ring-shaped, and enhancement could persist less than 2 weeks in some of the new lesions while some other new lesions would enhance during more than 4 weeks. Multiple studies demonstrated significant intrapatient variation in the frequency of gadolinium-enhancing lesions during the course of disease.3,7,44-47⇓⇓⇓⇓⇓ Taken together, these findings may suggest that periods of heightened lesion activity lasting a few months (perhaps triggered by an exogenous event such as infection by a microorganism) alternate with periods of relative quiescence. Our results show that within such outbursts of lesion activity, however, considerable heterogeneity in the activity of individual lesions may exist. The complexities of individual lesion behavior remain largely to be explored.
Acknowledgments
Supported in part by the NIH (R01 NS35142; P41 RR13218–01) and by the Foundation for Neurologic Diseases (C.R.G.G.). F.C. was supported in part by a grant from the Philips Foundation.
Acknowledgment
The authors thank Marianna Jakab for technical assistance; Drs. Christian Confavreux, Marc Hermier, and Chahin Pachai for critical revision of the manuscript; and Nancy Drinan for editorial comment.
- Received December 10, 2001.
- Accepted October 17, 2002.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.