OCR Text |
Show { oumal o{ Clinicll' Nellro- 0l'htha( moloxy 11( 3): 205- 212. 1991. rc] 1991 Raven Press. Ltd.. New York Pattern Visually Evoked Cortical Potential and Magnetic Resonance Imaging in Optic Neuritis Yonetsugu Kakisu, M. D., Emiko Adachi- Usami, M. D., and Naoya Fujimoto, M. D. We studied the extent of optic nerve lesion with magnetic resonance imaging ( MRI) in 24 patients with optic neuritis and compared it with visually evoked cortical potentials ( VECP). Patients were divided into acute and chronic stage groups according to the duration of the disease from its onset. The short T. inversion recovery ( STIR) mode of the MRI was selected for study. In the acute stage group, abnormally high signals of optic nerve on MRI were demonstrated in 11 of 12 eyes, and the amplitude and peak latency of pattern VECP were unrelated to the extent of the optic nerve lesion visualized by MRI. Conversely, in the chronic stage group, the abnormal MRI findings were exhibited in 19 of 23 eyes, and the relationship between the measurements was statistically significant. This difference in the optic nerve lesion observed by MRI between the acute stage and the chronic stage of optic neuritis may indicate pathophysiologic differences in their origins. Key Words: Pattern VECP- MRI- STIR mode- Optic neuritis. From the Department of Ophthalmology, School of Medi-cine, Chiba University, Chiba, Japan. . Address correspondence and reprint requests to Dr. Kaklsu at Department of Ophthalmo~ ogy, School of Medicine, Chiba University, Inohana 1- 8- 1, Chlba, 280 Japan. 205 The P- lOO peak latency of pattern reversal visually evoked cortical potentials ( VECP) is delayed in patients with optic neuritis as a result of demyelination ( 1,2). However, the delayed response cannot be explained solely by this demyelination ( 3). In addition, Halliday has reported that the delayed peak latency of the pattern VECP is retained, but its amplitude parallels the onset and subsequent recovery of the visual impairment at the time of the attack in demyelinating disease ( 4). Magnetic resonance imaging ( MRI) has been used to evaluate the involvement of the visual pathway in demyelinating disease, when X- ray computed tomography ( CT) has failed to determine positive information ( 5- 7). Magnetic resonance imaging, as represented by Tl and T2 relaxation times can characterize ocular tissue. It is difficult, however, to visualize the optic nerve under the condition of Tl- and T2- weighted images because of the artifactual chemical shift and the strong signal from orbital fat. In 1986, Miller and associates reported that the short T( inversion recovery mode ( STIR) on MRI suppresses the orbital fat signal, allowing visualization of the optic nerve. The authors indicated that optic neuritis was characterized by a high signal on MRI with use of the STIR mode ( 8). They also demonstrated that the extent of the optic nerve lesion in optic neuritis was related to the delayed abnormality of the pattern VECP ( 9). Such clinical results were not quantitatively evaluated. Therefore, we measured the extent of the high signal of the optic nerve lesion on MRI in patients with optic neuritis, and compared it with both peak latency and amplitude of the pattern VECP. SUBJECTS AND METHODS We examined 24 ( 7 men and 17 women) patients with optic neuritis, including 13 cases of unknown 206 Y. KAKISU ET AL. origin ( 10 unilateral, 3 bilateral), 10 with multiple sclerosis ( 3 unilateral, 7 bilateral), and 1 with acute dissemina ted encepha lornyelopa thy ( bilateral). The investigative nature of the procedure was thoroughly explained to each patient, and informed consent was obtained before the tests were done. The patients' mean age was 30.4 years ( range, 3- 50 years). A total of 35 eyes were examined. The MRI and pattern VECP were performed on the same or successive days. In the acute stage, the pattern VECP response is diminished by dysfunctional nerve conduction, and in the chronic stage, the peak latency of the pattern VECP is delayed without amplitude attenuation in optic neuritis ( 4). In addition, the abnormal MRI signals in the acute and chronic central nervous system lesions arise in different ways: Edema occurs in the former, and gliosis takes place in the latter ( 10). Therefore, to evaluate the data and clarify the pathophysiologic difference between the two stages, the patients were divided into two groups according to the duration of the active stage of the disease: less than 4 months in the acute stage group and longer than 4 months in the chronic stage group. To visualize specifically the optic nerve lesion by MRI ( Picker Inst., 0.5T, superconducting), we selected the STIR mode ( inversion time: 100 msec; repetition time: 1,500 msec; echo time: 40 msec) that suppresses the orbital fat signal and increases the signal of the optic nerve. In our previous study ( 11), we found that the intensity of MR images of the optic nerve and extraocular muscles were similar, as measured with Tl and T2 relaxation times ( optic nerve: Tl, 450 msec, T2, 58 msec; extraocular muscle: Tl, 500 msec, T2, 52 msec). Figure 1 shows the STIR image of a normal volunteer. The signals from optic nerve and extraocular muscles are distinguishable from those of orbital fat. According to the method of Miller et aI., the extent of the lesion by MRI was estimated as the number of coronal slices in which an abnormality was seen ( 9). With this procedure, either 8 or 12 serial slices of 5 mm thickness were examined in the coronal plane in all cases and in the axial plane in several cases. Abnormally high signals of optic nerve on MRI were assessed by at least two ophthalmologists and were evaluated by the following criteria: ( a) a clear difference in signals between the right and the left optic nerves; ( b) when similar high signals were found in both optic nerves, the signals were compared with those of the extraocular muscles; ( c) if the judgments of the observers were different, we considered the signal " '.',,, 11. M,. 3. 1991 FIG. 1. Coronal section of the normal brain by short T I inversion recovery ( 100/ 1,500/ 40) mode on magnetic resonance imaging. Arrows indicate the optic nerve. as normal. The MRI examination was performed two or three times in five of the 24 cases. The T2weighted image on MRI was selected to identify the lesions in the brain. Checkerboard pattern reversal VECPs were recorded as the patients fixated on the center of a television monitor from a viewing distance of 1.7 m. The temporal frequency was 3 and 12 reversalsl second, the mean luminance was 39 cd/ m2, and the contrast was 80%. The visual field was 11 x 7 degrees, and the check size was 28 ( 3 reversals/ second, transient) or 14 minutes ( 12 reversals/ second, steady- state) of arc. The peak latency and amplitude of the P- 100 component and the amplitude of the steady- state pattern VECP were measured. The abnormally delayed peak latency of the pattern VECP was defined when it was beyond the mean peak latency of normal subjects + 2 SD. In Figs. 2 and 3, the dotted line ( lower right) indicates the upper limit of the P- lOO peak latency of the pattern VECP in our laboratory ( mean + 2 SD, 122 msec). PATTERN VECP AND MRI IN OPTIC NEURITIS 207 • 0 • • • -• 0 I I I I 2 4 6 IS > 0:; "=~ E .. 10 Q. U... > Q. ~ ".. v'"; 0 0 •! •• 1 0 o 5 >- • .~. ~:; 0 I oe • o 05 • M • o 01 • btent 01 OptiC Nerve Lesion on MRt ( number of slice) non- L ' fco. dable r Extent of Optic Nerve Lesion on MRI ( number of slice) • • o 180 o :; 10 ~'" Q E .. l> - • Q. l> 0 Q. U .>.. • • • Q. • • • • I I I 0 4 oI ~ 160 Ii > u c .' 3" 140 "" .. Q'". g ,;: Q. 120 u... > Q. 100 • • •• • • Extent of OptiC Nerve LeSIon on MRI ( number of shee) Extent of Optic Nerve Lesion on MRI ( number of slicel FIG. 2. The relationship between visual acuity and pattern reversal visually evoked cortical potential ( PVECP) and the number of coronal slices of serial section of the optic nerve lesion on magnetic resonance imaging ( MRI) in the acute stage of optic neuritis. Open circles indicate mUltiple sclerosis; closed circles. unknown origin; open triangles. acute disseminated encephalomyelopathy. We employed chi- square analysis to find the statistical difference between the MRI and VECP abnormalities, and then assessed the correlation coefficient with the p value between the extent of the optic nerve lesions on STIR mode and the amplitude and P- 100 peak latency of the pattern VECP. RESULT Acute Stage Group In 7 of 8 eyes with optic neuritis of unknown origin and in all 4 eyes with acute disseminated encephalomyelopathy and multiple sclerosis, the high signals of the optic nerve were seen with use of the STIR mode on MRI. In comparison, an ab-normally enlarged optic nerve demonstrated by CT was found in only 5 of 10 eyes. With regard to the brain lesion, 3 of 8 cases with lesions of unknown origin and all 3 cases with acute disseminated encephalomyelopathy and multiple sclerosis had high signals from the white matter in the brain on T2- weighted image by MRI, while in only 1 of 11 cases were the abnormal findings on CT found. The differences were not statistically significant between the abnormal findings on the STIR mode of the MRI and the peak latency of the patt':' tn VECP using chi- square analysis. Figure 2 demonstrates the relationship between visual acuity ( left upper), steady- state ( 12 rev/ second) pattern VECP amplitude ( right upper), transient pattern VECP P- 100 amplitude ( left lower), and pattern VECP P- 100 peak latency ( right I Oill Neuro- ophthalmol, Vol. 11, No. 3, 1991 208 Y. KAKlSU ET AL. o o o • o Ell 0 024 6 Extent of Optic Nerve LeSion on MRI ( number of slice) o • • non- I rtcordable t- 15 >" .'."~. . E. 10 ... u... .>.... ~ ~ v; , ! o o • o 8 o co • • o 4 6 Extent of Optic Nerve Lesion on MRI ( number of shce) I 0 o 5 >- .~. ~:> 0 I o 05 001 , 0 180 • > 10 " • ~.. E 0 .. 0 ..~. ... 0 . u>.. • 00 e• ... & 0 • 0 • ! ! 8,3 9 , o 2 4 6 Extent of Optic Nerve LeSion on MRI ( number of shcel • v.. 160 e ~ 0 0 • ~ 140 0 ~ ~ 0 0 .... • ... • • ~ t 120 - ----- --~- - --- Q..------ .~.. 0 •• 100 6 E. tent 01 Optic Nerve lesion on MRI ( number 01 slice) FIG. 3. The relationship between visual acuity and pattern reversal visually evoked cortical potential ( PVECP) and the number of coronal slices of serial section of the optic nerve lesion on magnetic resonance imaging ( MRI) in the chronic stage of optic neuritis. Open circles indicate multiple sclerosis; closed circles. unknown origin. lower) and the number of serial coronal MRI slices with an abnormally high signal from the optic nerve in the acute stage of optic neuritis. The visual acuity at MRI measurement was less than 0.4 in 10 of 12 eyes. The P- lOO peak latency of the pattern VECP was abnormally delayed or nonmeasurable in 8 of 12 eyes in the acute stage group. The MRI showed high signals of the optic nerve lesion in 11 of 12 eyes. In all patients with multiple sclerosis, both the pattern VECP and MRI findings were abnormal. If the P- lOO peak latency of the pattern VECP was prolonged, then the extent of the optic nerve lesion as observed by MRI tended to be longer, although the difference was not statistically significant ( p < 0.1, r = 0.5). The visual acuity and amplitude of transient and steady- state pattern VECP at the acute stage was unrelated to the extent of the optic nerve lesion exhibited by MRI ( transient, r = - 0.24; steady- state, r = - 0.12). Chronic Stage Group In 6 of 8 eyes with optic neuritis of unknown origin and in 13 of 15 eyes with multiple sclerosis, the high signals of the optic nerve were seen by STIR mode on MRI, while an enlarged optic nerve was not demonstrated by CT in any of the 15 eyes in the acute stage group. In 1 of 5 cases of unknown origin and in 8 of 10 cases with multiple sclerosis, the high signals of the white matter in the brain were seen on the T2- weighted image by MRI, while in only 2 of 8 cases were abnormal findings on CT found in the brain. There was no J ( 1111 J,,:' PATTERN VECP AND MRI IN OPTIC NEURITIS 209 statistical significance between the abnormal findings of STIR mode on MRI and the pattern VECP. Figure 3 demonstrated the relationship between visual acuity ( left upper), steady- state ( 12 revl second) pattern VECP amplitude ( right upper), transient ( 3 rev/ second) pattern VECP P- lOO amplitude ( left lower), and transient pattern VECP P- lOO peak latency ( right lower) and the extent of the optic nerve lesion on MRI at the chronic stage of optic neuritis. The P- lOO peak latency of pattern VECP was delayed or nonmeasurable in 19 of 23 eyes in the chronic stage. The extent of the optic nerve lesion and the pattern VECP abnormalities were significantly correlated ( p < 0.005, r = 0.68). The extent of the optic nerve lesion and the amplitude of the transient and steady- state pattern VECP also were statistically significant ( transient, p < 0.025, r = - 0.46; steady- state, p < 0.005, r = - 0.65). However, the visual acuity was not related to the extent of the chronic stage optic nerve lesion on MRI. Two representative cases from our series are briefly described. Case 1 On January 26, 1989, a 44- year- old woman noticed a visual disturbance in the right eye. Visual acuity in the right eye was 0.02 and it was 1.5 in the left. The right optic disc appeared normal. Goldmann Visual Perimetry of the right visual field showed a central scotoma. The pattern VECP was extinguished in the right eye. The visual function of the left eye was normal. On MRI the right optic nerve demonstrated a large high signal extending 10 mm from the eyeball to intracranial optic nerve. The left nerve appeared normal ( Fig. 4, upper right, upper left, lower left). As seen in Fig. 4 ( lower right), the MRI signals of both optic nerves were not different from those of the extraocular muscles, and both optic nerves were the same size. Therefore, this portion of the right optic L FIG. 4. A 44- year- old woman ( Case 1) with right optic. neuritis of unkno~ n origin is seen by short T, inversion recove~ y mode on magnetic resonance Imaging. Arrows indicate the high signal of the optiC nerve. I Clin Neuro- ophthalmol. Vol. 11. No. 3. 1991 210 Y. KAKISU ET AL. nerve was estimated as normal. The length of the high signal of the right optic nerve extended 6 coronal slices ( 6 slices x 5 mm = 30 mm). Several high signals under the T2- weighted mode on MRI were demonstrated from the periventricular white matter. At this time the visual acuity in the right eye was light perception. The pattern VECP for the same eye was extinguished. She had no abnormal neurophysiologic signs, and her visual acuity recovered at the end of March. We characterized this case as optic neuritis of unknown origin in our study. Case 2 A 30- year- old woman was diagnosed in 1978 as having myasthenia gravis. After 2 years, a thymectomy was done. Since 1985, she had noticed at several times an abnormal sensory disturbance in both arms and legs. In 1986, the patient developed bilateral optic neuritis, which later improved. Finally, she was diagnosed as having multiple scle-rosis, for which she underwent steroid therapy, and her condition improved. In February 1989, right optic neuritis developed; the MRI was done 1 month later. At this time, visual acuity in the right eye was no light perception, and it was 1.2 in the left eye. The pattern VECP was not identified in the right eye and the P- lOO peak latency of pattern VECP was delayed to 123 msec in the left. The extent of the high signal of the right optic nerve occupied 7 coronal slices, which were estimated to be 35 mm, and 6 coronal slices ( 30 mm) in the left optic nerve ( Fig. 5). There were several high signals near periventricular white matter. A month later, the extent of the high signal in both optic nerves diminished, although the visual acuity remained as light perception. DISCUSSION Tl- and T2- weighted images are most frequently used in MRI. Sequences of the Tl image select short spin echo mode ( echo time: less than 40 FIG. 5. A 30- year- old woman ( Case 2) wltn Ollateral ~ PtIC neUritis resulting from multIple scl. eroslS IS seen by short T, recovery mode on magnetic resonance imaging. Arrows indicate the high Signal of the optic nerves. 1(. 11: . PATTERN VECP AND MRI IN OPTIC NEURITIS 211 msec; repetition time: less than 500 msec) and inversion recovery mode ( inversion time: 400 msec; repetition time: more than 1,200 msec). T2weighted images select long spin echo mode ( echo time: > 80 msec; repetition time: > 1,200 msec). The orbital image on these sequences is almost the same, since Tl and T2 relaxation times of orbital fat are shorter than for other orbital tissues. The signals of orbital fat are higher ( i. e., light) than other orbital tissues ( i. e., dark). In 1985, Bydder and Young introduced a new technique to suppress the signals of fat on MRI ( 12); the method was called short T( inversion recovery ( STIR) mode. Their idea was based on the following facts: To let the signals of the orbital fat become small, it is necessary to shorten the inversion time ( less than 150 msec) as the MRI signal of the fat is close to zero. However, because the signals of the optic nerve and extraocular muscle are high against those of fat, they can be seen as something like a white rope in a dark area. If the inflammation of the optic nerve occurs, long Tl and T2 relaxation times make the optic nerve brighter. Miller and associates demonstrated the high signal of the optic nerve lesion in 84% of 37 patients with optic neuritis, using the STIR mode with MRI ( 9). Our present study also demonstrated results similar to theirs and to our previous study ( 13). Guy and associates, however, detected abnormal findings in about 7 of 13 patients ( 54%), using the STIR mode with gadolinium- DTPA ( 14). This difference may have been due to the difference in the MRI equipment and scanning methods. We further showed a higher abnormal rate on MRI in the acute stage than in the chronic stage. It has been thought that in the acute stage of multiple sclerosis, the plaques of the white matter in the brain show edema and cellular infiltration, and then become smaller lesions as gliosis and demyelination are found in the chronic stage ( 10,15). Considering those pathologic findings, edema and demyelination appeared to be related to the detection rate of abnormality on MRI. Miller and associates found that slow or poor visual recovery was associated with more extensive lesions, but the authors did not describe the size of MRI lesions and correlate that to the delay of VECPs ( 9). We determined that, in the chronic stage, the decrease of the amplitude and the delay of the P- 100 peak latency of pattern VECP were correlated to the length of the optic nerve lesion. According to McDonald ( 16), a 1O- mm len~ th of ~ emyelination of the optic nerve was aSSOCiated With a 25- msec delay of the P- lOO peak latency of the pattern VECP. In our series, 1 patient in the chronic stage group had 4 coronal serial slices demonstrating a high signal in the involved optic nerve. However, the P- 100 peak latency of the pattern VECP was 130 msec, which was slightly delayed. There is no information on how long a segment of demyelination must be to produce an abnormally high signal in a 5- mm thick coronal section on MRI. In our series of MRI findings, we have not observed skip areas. Therefore, serial sections approximated the total length of involved nerve, and the consecutive serial section represented 20 mm. In the chronic stage, the high signal of the optic nerve on STIR sequence may have indicated not only demyelination but also gliosis. This close correlation between the amplitude and the P- 100 peak latency was not observed in patients in the acute stage group. The high signal of the optic nerve found in the acute stage might represent edema and not demyelination. In future studies, the development of the MRI technique to differentiate the composition of lesions such as edema, demyelination, and gliosis would be of great value. We agree with Miller and his associates that pattern VECP and STIR sequence on MRI are complementary in the evaluation of optic neuritis. Acknowledgment: This study was supported in part by grant 63480392 from the Ministry of Education, Japan. We wish to acknowledge Maxine Cere, M. S., for editing the manuscript. REFERENCES 1. Halliday AM, McDonald WI, Mushin J. Visual evoked response in diagnosis of multiple sclerosis. Br Med / 1973; 4: 661- 4. 2. Kakisu Y, Adachi- Usami E, Mizota E. Pattern VECPs in optic neuritis caused by multiple sclerosis. Acta Soc Ophthalmol/ pn 1987; 91: 230- 4. 3. Plant GT, Hess RF. The electrophysiological assessment of optic neuritis. In: Hess RF, Plant GT, eds. Optic neuritis. Cambridge: Cambridge University Press, 1986: 192- 229. 4. Halliday AM. The visual evoked potential in the investigation of diseases of the optic nerve. In: Halliday AM, ed. Euoked potentials in clinical testing. Edinburgh: Churchill Livingstone, 1982: 187- 234. 5. Gebarski 55, Gabrielsen TO, Gilman 5, et al. The initial diagnosis of multiple sclerosis: clinical impact of magnetic resonance imaging. Ann Neuro/ 1985; 17: 469- 74. 6. Johns K, Lavin P, Elliot JH, et al. Magnetic resonance imaging of the brain in isolated optic neuritis. Arch Ophthalmol 1986; 104: 1486- 8. 7. Kakisu Y, Adachi- Usami E, Kojima 5, Hirayama K. Magnetic resonance imaging ( MRI) in the diagnosis of optic neuritis and neuropathy. Acta Soc Ophthalmol ] pn 1989; 93: 281- 6. 8. Miller DH, Johnson G, McDonald WI, et al. Detection of I Clin Neuro- ophthalmol, Vol. 11, No. 3, 1991 212 Y. KAKISU ET AL. optic nerve lesions in optic neuritis with magnetic resonance imaging. umeet 1986; 1: 1490-- 1. 9. Miller DH, Newton MR, van der Poel IC, et at. Magnetic resonance imaging of the optic nerve in optic neuritis. Neurology 1988; 38: 175- 9. 10. Ormerod IEC, Miller DH, McDonald WI, et aJ. The role of NMR imaging in the assessment of multiple sclerosis and isolated neurological lesions. Brain 1987; 110: 1579- 1616. 11. Kakisu Y. Application of MRl in ophthalmology. Ganka 1988; 30: 915- 20, 1988. 12. Bydder BM, Young IR. MR imaging: clinical use of the inversion recovery sequence. J Comput Assist TOl/ lOg 1985; 9: 659- 75. 13. Kakisu Y, Adachi- Usami E, Fujimoto N. MRllesions of the optic nerves in optic neuritis. Acta Soc Ophthalmol ' pn 1990; 94: 673- 8. 14. Guy J, Mancuso A, Quisling RG, et al. Gadolinium- DTPAenhanced magnetic resonance imaging in optic neuropathies. Ophthalmology 1990; 97: 592- 600. 15. Adams CWM. The general pathology of multiple sclerosis: morphological and chemical aspects of the lesions. In: Hallpike IF, Adams CWM, Tourtellotte WW, eds. Multiple sclerosis: pathology, diagnosis and management. London: Chapman and Hall, 1983: 203- 40. 16. McDonald WI. Pathophysiology of conduction in central nerve fibres. In: Desmedt JE, ed. Visual evoked potentials in man: new developments. Oxford: Clarendon Press, 1977: 42737. |