Subclinical Visual Involvement in Multiple Sclerosis: A Study by MRI, VEPs, Frequency-Doubling Perimetry, Standard Perimetry, and Contrast Sensitivity

Abstract

purpose. To evaluate the effectiveness of visual evoked potentials (VEPs), frequency-doubling perimetry (FDP), standard achromatic perimetry (SAP), contrast sensitivity (CS) test, and magnetic resonance imaging (MRI), isolated or in combination, in detecting subclinical impairment of visual function in multiple sclerosis (MS).

methods. Twenty-two eyes of 11 patients affected by clinically definite MS, without a history of optic neuritis and asymptomatic for visual disturbances, underwent full ophthalmic examination and, in addition, VEPs, FDP, SAP, CS, and MRI. Abnormal results were taken to be as follows: for VEPs, a P100 latency >115 ms; for FDP, abnormal mean deviation (MD) or pattern SD (PSD); for SAP, abnormal MD or PSD; for CS, abnormal CS at one spatial frequency, at least; and for MRI, evidence of at least one demyelinating plaque along the visual pathway.

results. VEPs showed abnormal results in 12 eyes (54.4%), FDP in 11 (50%), SAP in 14 (63.6%), CS in 17 (77.1%), and MRI in 16 (72.7%). In only two (9.1%) eyes of the same patient was no abnormality found. No single test detected all the abnormal eyes. Four (18.2%) eyes had pure optic nerve involvement and the remaining 16 (72.7%) had both pre- and postchiasmal involvement.

conclusions. In patients affected by clinically definite MS without history of optic neuritis and no visual symptoms, there is a large prevalence of visual pathway involvement that can be diagnosed only by performing multiple tests. The comparison of the tests is also useful to detect the presence of multiple lesions in the same patient.

Multiple sclerosis (MS) is often associated with involvement of the visual pathway that can lead to clinically evident manifestations, such as optic neuritis, nystagmus, and diplopia, ¹ and to more frequent subclinical manifestations. In some cases, the patient reports blurred vision even if visual acuity is normal. In other cases, no ocular symptoms are reported, but specific examinations can reveal subclinical abnormalities. The sensitivity of MRI to brain abnormalities in MS is higher than any other instrumental examination. ^{2 3} In addition, the specific fat-suppressed SE images or short tau inversion recovery (STIR) can detect lesions in the optic nerve in most patients with MS who have symptomatic acute optic neuritis. ^{4 5}  

Alteration of visual evoked potential (VEP) latencies during pattern stimulation is considered one of the most characteristic electrophysiological signs in patients with MS. Such alterations are present almost invariably in subjects affected by optic neuritis ⁶ and, in a percentage that varies from 35% to 93%, also in patients without symptoms and signs of visual system impairment. ⁷ Subjective and psychophysical tests, such as contrast sensitivity (CS) test, visual field examination by standard achromatic perimetry (SAP), color vision test and best-corrected visual acuity (BCVA) measurement, have been reported to be useful, but have been judged more or less sensitive according to different studies. ^{8 9 10 11}  

Frequency-doubling perimetry (FDP), first developed as a diagnostic test for glaucoma, is thought to isolate the function of a specific subtype of magnocellular retinal ganglion cells, the My cells, and then to reveal the presence of specific damage to these cells. FDP has been studied also in neuro-ophthalmic disorders, in comparison to standard achromatic perimetry, with similar results in terms of sensitivity and specificity. ^{12 13} Furthermore, FDP shows a general depression in the midperipheral area of the visual fields tested in patients with resolved optic neuritis. ¹⁴  

The purpose of this study was to evaluate the effectiveness of VEPs, FDP, SAP, CS, and MRI in detecting subclinical abnormalities in a population of visually asymptomatic patients affected by clinically definite MS. 

Materials and Methods

From January to October, 2002, 11 patients with clinically definite MS, according to Poser’s criteria ¹⁵ (22 eyes), all without previous optic neuritis and with no visual symptoms, were enrolled in the study. All the patients were recruited at the Department of Neurologic and Psychiatric Sciences of the University of Bari. Of the 11 patients, 3 were male and 8 female (mean age, 34.5 ± 10.1 years; range, 17–55). Ten of them had a relapsing–remitting and one a primary progressive MS course. In all the patients, the disease was in remission. No patient was affected by any general disease apart from MS. All the patients underwent full ophthalmic examination, including best-corrected visual acuity (BCVA) measurement, biomicroscopy, applanation tonometry, and fundus examination after pupil dilation. VEP pattern reversal was performed (Sirius Galileo; Esaote Biomedica, Florence, Italy). A P100 latency >115 ms (i.e., 2 SD above the mean) was considered abnormal. 

Standard achromatic perimetry (SAP) was performed by means of the Humphrey Field Analyzer 750 (model 750; Carl Zeiss Meditec, Dublin, CA) using the 30-2 program with the Swedish interactive threshold algorithm (SITA) standard strategy. Two consecutive visual field examinations were performed on each patient, but only the second was evaluated for the purpose of the study and only if reliability indices were normal. Criteria for abnormality were mean deviation (MD) or pattern standard deviation (PSD) labeled abnormal by the instrument’s software. 

FDP was performed by means of frequency-doubling technology (Carl Zeiss Meditec; Welch Allyn, Skaneateles, NY), with N-30 threshold. With this test too, only the second examination was considered for the purpose of the study, if reliability indices were normal. Criteria for abnormality were also MD or PSD labeled abnormal by the instrument’s software. 

CS evaluation was performed with the normalized version of the CSV-1000E Chart (VectorVision, Dayton, OH), proposed by Boxer Wachler and Krueger. ¹⁶ For this chart, normalized ratios were obtained by dividing each log CS by the population mean obtained for each spatial frequency shown (3, 6, 12, and 18 cyc/deg), rounded to the nearest 0.05 cyc/deg. The normalized CS mean is 1.0 (Fig. 1) . Sensitivities greater than 1.0 represent the proportion above the normalized CS mean, and those <1.0 are below the mean. All the values of CS falling outside the gray area delimiting the normal range on the specific chart (Fig. 2)were considered abnormal; in addition, mean and SD log CS of each spatial frequency was calculated. 

MRIs of orbits, brain, and spinal cord were performed in all patients with MS within 4 weeks of the ophthalmic examination, on a 1.5-Tesla scanner (Siemens Magnetom; Erlangen, Germany). T₁-weighted spin-echo (repetition time [TR], 2200 ms; echo time [TE], 80 ms), T₂-weighted spin-echo (TR 600 ms, TE 15 ms) and orbital coronal fat-suppressed fast spin-echo (short tau inversion recovery [STIR] TR 1560 ms, TE 50 ms) images were obtained, with a 256 × 256 matrix and a 5-mm slice thickness. Gd-DTPA (gadolinium-diethylenetriamine pentaacetic acid) was given intravenously in a dose of 0.2 mM/kg. Presence of Gd-DTPA-enhancing lesions was determined by an experienced neuroradiologist, blind with regard to clinical data, approximately 15 to 20 minutes after contrast injection. 

The research adhered to the tenets of the Declaration of Helsinki for human studies. 

All data were recorded on a spreadsheet and analyzed on computer (Excel software, version 9.0; Microsoft, Redmond, WA, and Instat software; GraphPad Software Inc., San Diego, CA). Continuous variables were analyzed with the Student’s t-test to detect differences between two groups, and one-way ANOVA for multiple comparisons between groups. P < 0.05 was considered statistically significant. 

Results

Ophthalmic examination produced normal findings in all the patients. BCVA was 20/20 (Snellen acuity) in all cases. 

Analytic data of instrumental examinations are reported in Table 1 . MRI showed no involvement of visual pathway in three cases (patients 1, 5, and 9). Sixteen (72.7%) eyes of eight patients showed visual pathway localization: involvement of the posterior optic radiation (OR) in five cases (patients 2, 3, 4, 10, and 11) and of the posterior limb of the internal capsule (IC) in two cases (patients 7 and 8). Patient 6 had involvement of both posterior optic radiation and posterior limb of the internal capsule (Table 1) . VEPs showed prolonged P100 latency in 12 (54.4%) eyes. Four patients had bilateral involvement (patients 2, 7, 8, and 10), and four had unilateral involvement (patients 3, 4, 6, and 11; see Table 1 ). FDP showed abnormal MD in 11 (50%) eyes and abnormal PSD in 5 (22.7%). Mean MD was −2.7 ± 3.6 dB, and mean PSD was 4.77 ± 1.81 dB. SAP showed abnormal MD in 14 (63.6%) eyes and abnormal PSD in 4 (18.1%). Mean MD was −2.88 ± 2 dB, and mean PSD was 2.26 ± 1.17. Almost all visual field losses observed in this study were diffuse, as shown by the large prevalence of MD alterations, both with FDP and SAP. Among the eyes with any defect, four had a central depression up to 10 central degrees, and four a central depression extended to 20 central degrees. Two had a dense partial right homonymous hemianopia, and two had a dense partial left homonymous hemianopia. One patient had, only in the left eye, a small but deep central scotoma, with sparing of the fixation point (Table 1) . 

Overall, CS was abnormal for at least one spatial frequency in 17 (77.1%) eyes. Specifically for each spatial frequency, CS was normal in all eyes at 3 cyc/deg but was abnormal in 2 (9.1%) at 6 cyc/deg, in 14 (63.6%) at 12 cyc/deg, and in 13 (60%) at 18 cyc/deg. In addition, mean log CS at 3, 6, 12, and 18 cyc/deg was 0.99 ± 0.13, 0.92 ± 0.1, 0.67 ± 0.2, and 0.67 ± 0.25, respectively (P < 0.0001 at one-way ANOVA test), showing a specific reduction of CS for medium-high spatial frequencies. 

Discussion

Many studies have shown the presence of subclinical alterations of the visual system in multiple sclerosis, but there is little agreement about the prevalence of these abnormalities or about which examination is most sensitive to them. In patients affected by definite MS without clinically evident visual involvement, abnormalities were found by VEPs in 68% to 100% of cases, ^{17 18} CS in 62% to 100%, ^{19 20 21} and SAP in 48% to 75%. ^{8 22}  

In addition, in several studies, two or more tests were performed on the same sample of patients: Van Diemen et al. ⁹ found 81.8% had abnormal VEPs and 72.7% had abnormal CS (the combination of the two test found abnormalities in 90.9% of the sample). Kupersmith et al. ¹⁰ found only 38% abnormal VEPs (60%, considering only cases of clinically definite MS) and 78% abnormal CS. Finally, Sanders et al. ¹¹ found abnormalities only in 5% by SAP, 14% by color vision test, and 27% by CS. 

As far as we are aware, this is the first study to evaluate a population of visually asymptomatic patients with MS by means of five different examinations, both anatomic (MRI) and functional (VEPs, FDP, SAP, and CS test). Considering the outcome of the study, several items should be discussed. 

A large prevalence of visual pathway involvement has been found in these patients. It is notable that the only patient (patient 9; Table 1 ) with no abnormalities at all, had spinal cord but no brain localizations of the disease. This finding seems to suggest that almost all the patients with MS with any brain localization could also have visual pathway involvement. By comparing the results of all the examination, we found that 4 eyes (patients 1 and 9) had pure optic nerve involvement, and the other 16 (72.7%) eyes had both pre- and postchiasmal involvement. 

No single examination detected all cases of visual involvement in our sample. This could be explained considering that even the most used tests are limited in different ways. In fact, the VEP P100 latency is the most diffuse and the parameter most often used to detect optic nerve involvement, but it is not very sensitive in the diagnosis of postchiasmal localizations. ²³ An example is provided in our sample, by the right eye of patient 3: VEPs latency was normal in the presence of homonymous left hemianopia and alterations of SAP, FDP, and the CS test. On the contrary, MRI is the gold standard examination for the detection of demyelinating plaques of the brain but, despite the use of specific orbital fat-suppressed STIR sequences, it is often not able to detect optic nerve involvement. In fact, in our sample, MRI found no optic nerve involvement; otherwise VEPs were abnormal in 12 eyes. 

The brain localizations detected by MRI were not always able to determine visual field defects. In fact, only two patients had typical (but not complete) homonymous hemianopia (patients 3 and 10; Table 1 ). According to Plant et al. ²³ several factors could explain this finding: first, in the optic radiation the fibers occupy a large area (not a small volume as in the optic nerve), so that a typical MS lesion—1 cm long × 5 mm in diameter—is not likely to damage a large proportion of these fibers; second, the lesions are oriented to venules, not to fiber tracts, and they may have little effect on posterior fibers that follow a curvilinear course; third, in the optic nerve, there are fibrous septa that may restrict expansion and then add a compressive element to the effect of demyelination, whereas in the cerebral hemisphere the potential for spread of edema is much greater. In fact, Plant et al. found cases of complete hemianopia only in patients with very large postchiasmal lesions. Such lesions were not present in our study, and this could also explain why these visual field defects were not symptomatic. 

VEPs P100 latency was not always able to detect all cases of optic nerve involvement. In fact, eyes with normal P100 latency had a clear optic nerve involvement as diagnosed by visual field central defect and abnormal CS test results (patient 4, right eye; patient 11, right and left eye; Table 1 ). 

In two patients (2 and 7) it was difficult to determine the seat of the lesion that caused the functional abnormalities. In fact, MRI showed bilateral postchiasmal involvement; therefore, abnormalities of VEPs and CS test results and also central visual field defects in patient 7 (could be considered a double hemianopia) could represent either concurrent optic nerve involvement or the consequence of the bilateral postchiasmal lesions. 

Among the single examinations, FDP showed the worst rate of abnormalities detected (50%); in addition, in no case it was the only test with abnormal results. Considering that FDP has been developed to study the function of a specific type of retinal ganglion magnocellular cell, the My cell, it is likely that these cells are not specifically affected in MS. In fact, Evangelou et al., ²⁴ studying the brains of eight patients who died of MS, found a relatively selective atrophy of the small neurons of the parvocellular layers in the lateral geniculate nucleus, thus confirming this hypothesis. 

As for CS, our study showed a specific involvement of higher spatial frequencies, especially of 12 and 18 cyc/deg. This finding is consistent with those in previous studies. The most frequent abnormality of CS that Sanders et al. ¹¹ found in 45 unaffected eyes of patients with MS, was a reduction of high spatial frequencies (spatial frequency range explored: 0.1 to 25.6 cyc/deg). In contrast, in a similar sample of patients, Kupersmith et al. ¹⁰ found that diffuse loss of CS was the most consistent pattern in eyes with any CS abnormality, but the spatial frequency they explored ranged from 0.2 to 6.4 cyc/deg, and thus higher frequencies were not tested. 

In conclusion, our study showed that, in patients affected by MS with no history of optic neuritis and no visual symptoms, there is a large prevalence of visual pathway involvement. In addition, there is not a specific examination that reveals all cases of visual pathway involvement. As discussed, this could depend on several factors: lack of a specific affected cellular type; lack of a specific visual function involved; and, more important, location, extension, and orientation of the plaques in relation to the course of the fibers. Nevertheless, the comparison of multiple examinations can be useful to detect nearly all cases of visual pathway involvement, to diagnose the presence of multiple visual pathway localizations and, in some cases, to understand which lesion is responsible for the functional abnormalities. Further studies are needed to confirm these findings in a larger series of patients, also including those with subclinical, MRI-documented, involvement of the anterior visual pathway. 

 Submitted for publication November 5, 2003; revised April 20, May 12, June 23, and July 2, 2004; accepted July 20, 2004.

 Disclosure: D. Sisto, None; M. Trojano, None; M. Vetrugno, None; T. Trabucco, None; G. Iliceto, None; C. Sborgia, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

The normalized CSV-1000E. Ratios can be written in patient’s charts, easily interpreted by the clinicians, and also used for calculating mean values in given populations of patients (reprinted, with permission, from Boxer Wachler BS, Krueger RR. Normalized contrast sensitivity values. J Refract Surg. 1998;14:463–466. © SLACK, Incorporated).

Figure 2.

 

Typical defects found in a patient. (A) Midperipheral defect of visual field shown by FDP. (B) Similar defect shown by standard perimetry. (C) Specific loss of medium-high spatial frequencies CS.

Table 1.

 

Analytic Data of the Examined Eyes

Patient/Eye	MRI Locations of Lesions	VEP (P100 Latency)	Perimetric Findings (SAP)	FDP			SAP			CS
								MD	PSD	MD	PSD	3 cyc/deg	6 cyc/deg	12 cyc/deg	18 cyc/deg
1/RE	Neg.	Neg.	Neg.	2.29	3.71	−0.11	1.74	1.1	0.95	0.6	0.85
1/LE		Neg.	Neg.	3.35	3.24	−1.07	1.57	1.1	0.9	0.7	0.85
2/RE	Left and right temporal lobe (OR)	>115 msec	Neg.	−1.58	3.8	−0.75	1.51	0.8	0.7	0.4	0.4
2/LE		>115 msec	Neg.	−1.5	3.94	−1.79	1.96	0.8	0.65	0.6	0.4
3/RE	Occipital horn of right lateral ventricle (OR)	Neg.	Homonymous left hemianopia	−7.67	7.12	−5.71	3.41	0.8	0.9	0.4	1.1
3/LE		>115 msec	Homonymous left hemianopia	−8.54	10.19	−6.49	5.38	1.0	0.9	0.4	0.4
4/RE	Occipital horn of right lateral ventricle (OR)	Neg.	Central 15°	−2.36	3.54	−2.82	1.65	1.1	1.05	0.8	0.55
4/LE		>115 msec	Central 15°	−4.13	4.52	−3.75	1.91	1.1	0.95	0.8	0.85
5/RE	Neg.	Neg.	Neg.	0.17	3.54	−0.3	1.38	1.15	1.05	1.1	0.95
5/LE		Neg.	Neg.	−0.23	3.3	−0.75	1.53	1.1	1.1	1.0	1.1
6/RE	Occipital horn of right lateral ventricle (OR) + posterior limb of right IC	>115 msec	Neg.	2.23	3.77	−0.96	1.53	1.0	1.05	0.6	0.4
6/LE		Neg.	Neg.	3.84	3.04	−0.72	1.46	1.15	1.05	0.8	0.4
7/RE	Posterior limb of right and left IC (OR)	>115 msec	Central 15°	−5.83	4.9	−3.13	1.85	1.0	0.9	0.7	0.55
7/LE		>115 msec	Central 15°	−4.17	6.58	−5.08	2.56	0.9	0.9	0.6	0.85
8/RE	Posterior limb of left IC (OR)	>115 msec	Central 20°	−7.29	6.77	−4.7	2.4	1.25	0.9	0.7	0.4
8/LE		>115 msec	Central 20°	−7.38	7.67	−4.69	1.62	0.8	0.9	0.9	0.55
9/RE	Neg.	>115 msec	Neg.	−3.81	4.47	−0.75	1.64	1.1	0.9	0.7	0.7
9/LE		Neg.	Small central scotoma	−3.97	3.5	−2.15	1.81	0.8	0.95	0.9	1.1
10/RE	Left temporal lobe (OR)	>115 msec	Homonimous right hemianopia	−5.11	4.42	−3.87	2.08	0.9	0.9	0.4	0.4
10/RE		>115 msec	Homonimous right hemianopia	−4.04	3.95	−6.44	5.73	0.9	0.9	0.8	0.9
11/RE	Left temporal lobe (OR)	Neg.	Central 20°	−2.23	5.42	−3.55	2.64	1.0	0.9	0.4	0.55
11/LE		Neg.	Central 20°	−1.53	3.76	−3.96	2.56	1.1	0.9	0.6	0.7

 Abnormal results are shown in bold. Neg., negative for locations of the visual pathway.