Drs. Kepecs and Boro contributed equally to this work.
Address correspondence and reprint requests to Dr. M.R. Kepecs at Dept. of Neurology, St. Luke's-Roosevelt Hospital, 1000 Tenth Ave., New York, NY 10019, U.S.A. E-mail: firstname.lastname@example.org
Summary: Purpose: Pharmacotherapy for photosensitive epilepsy is not always effective and is associated with well-recognized toxicities. Nonpharmacologic approaches to the management of photosensitive epilepsy have included the use of sunglasses of various types. Blue lenses have been shown to suppress the photoparoxysmal response more effectively than lenses of other colors with similar overall transmittances. Recently, cross-polarized glasses have shown promise. The axes of polarization of the two lenses of such glasses are perpendicular to one another. We tested the effect of combining the use of blue and cross-polarized lenses in three patients with photosensitive epilepsy.
Methods: We recorded the EEG response to photic stimulation, television screens, and computer monitors in three patients with photosensitive epilepsy. If photoparoxysmal responses were provoked in any of these scenarios, testing was repeated with the patient wearing nonpolarized, parallel-polarized, and blue cross-polarized sunglasses.
Results: One of our patients had clinical seizures that were inadequately suppressed with moderate doses of valproate (VPA) but completely suppressed with blue cross-polarized lenses. The second patient's photoparoxysmal response was suppressed by both parallel-polarized and blue cross-polarized glasses, whereas the third patient's photoparoxysmal response was not suppressed by either.
Conclusions: These preliminary data suggest that blue cross-polarized lenses may be useful in the treatment of photosensitive epilepsies and that their efficacy can be predicted in the EEG laboratory.
Nonpharmacologic approaches to the management of photosensitive epilepsy (PSE) have included the use of optical filters of various types. The effectiveness of sunglasses, in particular, is known to depend on both the overall transmittance and the color of the lenses. Capovilla et al. (1) reported that Z1 lenses, a particular type of blue lens with relatively low transmittance at the deep-red end of the visible spectrum, abolished the photoparoxysmal response (PPR) in 64 (77%) of 83 patients and attenuated it in 16 (19%).
Cross-polarized glasses also have shown promise. The axes of polarization of the two lenses of such glasses are perpendicular to one another. Jain et al. (2) described a 16-year-old patient with seizures provoked by stimuli such as watching television and flickering lights. By using glasses with lenses that had independently variable polarization axes, they found that the PPR was most effectively suppressed when the axes were at right angles. The glasses were least effective when the polarization axes were parallel. The glasses were well tolerated and prevented further seizures. Jain et al. (3) subsequently compared the effects of parallel and cross-polarized lenses in 19 patients. Both types of polarized glasses attenuated the PPR in 17 patients, and the cross-polarized glasses were more effective in 10.
We report the effect of combining the use of blue and cross-polarized lenses in three patients with PSE.
Each patient underwent a detailed clinical assessment and testing for photosensitivity. Photic stimulation was delivered with a standard xenon strobe lamp (XLTEK), delivering flashes through a pattern-free diffuser. Unless otherwise stated, 5-s trains of flashes at 1, 3, 6, 9, 12, 15, 18, and 21 Hz were presented with the lamp placed ∼30 cm from the nasion. Commercially available glasses with both lenses polarized at 180 degrees and custom cross-polarized glasses with the right lens polarized at 90 degrees and the left polarized at 180 degrees were used. The patients' eyes were closed during photic stimulation without glasses and open during photic stimulation with glasses. The patients also were exposed to television screens and computer and hand-held electronic screen games with video-EEG monitoring for a minimum of 5 minutes under each condition. When possible, the videotapes and electronic screen games that had provoked events outside of the laboratory were used. PPRs were scored if generalized spikes-and-waves or polyspikes-and-waves were present in response to the stimulus. More restricted forms of the PPR were not encountered in this study. We did not require that the PPR outlast the stimulus (4). All testing was done for clinical indications. The characteristics of the blue parallel and cross-polarized and gray cross-polarized lenses we used are described in Table 1.
Table 1. Lens characteristics
Manufacturer and model
Blue parallel and cross-polarized
Specialty Lens CR-39
43% (lens alone)
23–25% (lens + dye)
Additional tint with Optisource Nu-Chem Blue UV dye
Younger Optics NuPolar Gray 3
A 9-year-old boy had a seizure while watching his brother play a computer game. He initially saw lights and colors from the left side of his visual field. Subsequently, his head turned to the left, his right hand became clenched, and he began opening and closing the hand. This was followed by drooling from the right corner of his mouth; he then became lethargic and did not respond to verbal commands for several minutes. Interestingly, he later reported that he had similar episodes of seeing lights when playing computer games, as well as episodes of dizziness while watching movies. His father had a generalized tonic–clonic seizure 20 years before. Development, neurologic examination, and magnetic resonance imaging (MRI) of the brain were normal.
EEG revealed bilaterally independent polymorphic slowing in the posterior temporal regions without epileptiform features. A PPR was present during photic stimulation at 18, 21, and 25 Hz, as well as in response to viewing a television and a computer screen from 15 cm (Fig. 1A). Pink, gray, and brown sunglasses (both polarized and nonpolarized) failed to block the PPR. Valproate (VPA) was initiated at a dosage of 125 mg twice a day. An additional generalized seizure occurred while the patient was watching the Superbowl on television, with a VPA level of 35 μg/ml. The VPA dose was increased to 187.5 mg in the morning and 125 mg at night.
Additional testing showed that the PPR was present at multiple stimulation frequencies with the lamp placed 15 cm from the nasion (but not 30 cm) and in response to a television viewed from 15 cm, but not from 2 m. No PPRs were elicited when the patient played a computer game at distances of 5 and 30 cm. Gray cross-polarized glasses blocked the PPR induced by viewing a television from 15 cm, but not the PPRs induced by photic stimulation. By contrast, blue parallel-polarized lenses blocked the PPR induced by photic stimulation, but not the PPR induced by viewing a television from 15 cm. Blue cross-polarized glasses blocked the PPR induced both by photic stimulation and by close television viewing (Fig. 1B).
VPA was discontinued because it was judged that the results of the laboratory testing and concerns about the long-term side effects of antiepileptic medications (AEDs) justified a trial of nonpharmacologic therapy. The patient is now wearing blue cross-polarized glasses for all computer and television viewing. The glasses are well tolerated and are not associated with subjective eyestrain. During the 18 months of follow-up, he has had one subsequent seizure, which occurred while viewing a computer without his glasses. He has not reported any subjective visual phenomena or dizziness.
An 18-year-old girl had a single generalized tonic–clonic seizure while riding in a car at age 13 years. Before this event, she could often be found with her face centimeters from a television or computer monitor, in what her parents described as a trance-like state. Her development, neurologic examination, and MRI of the brain were normal. No family history of epilepsy was found.
At age 13 years, the baseline EEG was normal. PPRs were present in response to photic stimulation at a variety of frequencies, as well as in response to a television viewed at 30 cm. She appeared to be drawn to the television and demonstrated decreased responsiveness and rhythmic bobbing of the head during the period of stimulation. Nonpolarized and parallel-polarized sunglasses of various types were tried; only “Ray-Ban” polarized sunglasses abolished the PPRs induced by both photic stimulation and television viewing. The EEG was repeated when she was 18 years old. PPRs were present at multiple stimulation frequencies. No PPRs were present when the patient played a video game and viewed a television from a variety of distances. Both the patient's dark parallel-polarized sunglasses and blue cross-polarized glasses blocked the PPR induced by photic stimulation.
The patient was an outstanding student at the time of diagnosis. She was not treated with AEDs because of concerns about the cognitive and long-term side effects of AEDs and strong family preferences. Since her first EEG, the patient has worn dark parallel-polarized sunglasses for all computer and television viewing. Although she no longer appears to be drawn to television screens or computer monitors, she frequently moves her fingers in front of her eyes in a rhythmic manner. During the 5 years of follow-up, she has had no photoconvulsive or tonic–clonic seizures. She is now attending an Ivy League School.
A 16-year-old neurologically normal boy had a generalized seizure while viewing a malfunctioning, rhythmically flashing computer screen. His development, neurologic examination, and MRI of the brain were normal. No family history of epilepsy was noted.
Routine EEG was normal. PPRs were present at multiple stimulation frequencies with the stimulator placed at 15 cm, but not 30 cm. PPRs were present when a television was viewed from 15 cm. The patient returned to the laboratory for testing with sunglasses. This time, rare, spontaneous frontally maximal generalized 2.5- to 3-Hz spike–wave complexes were present. Both the patient's dark nonpolarized and blue cross-polarized glasses failed to block PPRs with the stimulator at 30 cm at stimulation frequencies of 9, 12, 15, and 18 Hz.
The patient was not treated with an AED because he had only a single seizure that was provoked by an unusual stimulus he is unlikely to encounter on a regular basis. He had had no further seizures at the time of last follow-up, 1 year after his initial presentation.
In summary, for the first patient, only blue cross-polarized glasses blocked the PPR in response to the full range of provocative stimuli that were offered. The second patient's PPR was suppressed by both dark parallel-polarized and blue cross-polarized glasses, whereas the third patient's PPR was not suppressed by either (Table 2).
Table 2. Summary of results
TV at 15 cm
IPS, intermittent photic stimulation; PPR, photoparoxysmal response; +, presence of PPRs; −, absence of PPRs; ND, stimulation under that condition was not delivered.
aSeveral nonpolarized and parallel-polarized sunglasses of various types were tried in the initial evaluation of patient 2 (at age 13 years). Only Ray-Ban polarized sunglasses abolished the PPRs induced by both photic stimulation and television viewing.
Gray cross-polarized glasses
Blue parallel-polarized glasses
Blue cross-polarized glasses
Patient's dark parallel-polarized glasses
Blue cross-polarized glasses
Patient's dark nonpolarized glasses
Blue cross-polarized glasses
AEDs that have been used in PSE with variable results in small trials include clonazepam (CZP) (5), lamotrigine (LTG) (6), levetiracetam (LEV) (7), and vigabatrin (VGB (8). In the largest trial to date, Harding and Jeavons (9) found that photosensitivity was abolished by sodium VPA in 40 of 65 patients.
Because of the potential side effects of pharmacologic treatment, and because pharmacologic treatment is not always effective, nonpharmacologic approaches to the management of PSE are desirable. The most basic, effective, and unfortunately often impractical method is to avoid provocative stimuli. For many patients, this would require unacceptable restrictions on their activities in school and the workplace and on their mobility.
Attenuation of the stimulus is an alternative approach. Covering one eye is often an effective strategy for managing a transient threat (10). These authors suggest that photosensitive patients should view television in a well-lit room from a distance of ≥2 m and should not approach the television. The use of adaptive temporal filters, which reprocess video images to render them less epileptogenic, has recently been explored (10). This technology is still under development and addresses only a subset of the provocative stimuli that a photosensitive patient encounters in daily life.
Our preliminary data suggest that blue cross-polarized glasses may prevent photogenic seizures in some patients in whom other types of glasses with similar overall transmittances are less effective, and that their effectiveness in preventing PPRs can be predicted in the EEG laboratory. The glasses are well tolerated and, compared with the cost of therapy with AEDs, relatively inexpensive. Our cost for the blue cross-polarized lenses was $190.00 (without frames, in the northeastern United States). In our area, a typical cost for 1 year of therapy with generic VPA at a dose of 125 mg TID is $514.65 (11). This does not include the cost of laboratory testing. We suggest that these data are sufficiently compelling to justify the prospective evaluation of blue cross-polarized glasses in a larger group of patients.
The mechanism by which cross-polarizing glasses may attenuate the PPR is unknown. Jain et al. (2) suggested that cross-polarized lenses may render the subject monocular to flickering polarized light in the environment. This hypothesis does not explain the effectiveness of cross-polarized lenses in cases in which they block the PPR induced by a nonpolarized stimulus, such as the xenon arc lamp used for intermittent photic stimulation (IPS) or a standard television screen. The fact that cross-polarized lenses sometimes attenuate a PPR induced by a nonpolarized stimulus when parallel-polarized lenses are ineffective suggests that binocular disparity with respect to the plane of polarization of the stimulus introduced by the cross-polarized lenses is a relevant factor. Binocular disparity with respect to other components of the stimulus is known to attenuate the PPR; Wilkins et al. (12) used a stereoscope to present horizontal and vertical gratings independently and found that patterns that fused in binocular vision were significantly more epileptogenic than were those that did not.
The difficulty is that humans cannot consciously identify the plane of polarization of a visual stimulus (13). Evidence exists, however, that under some circumstances, changes in the plane of polarization of a visual stimulus result in changes in patterns of activation of the human retina and subsequent stages of visual processing. An example is the entoptic phenomenon of “Haidinger's brushes,” where “brushes” radiating from the point of fixation are seen when a subject views a rotating polarized stimulus (14). The brushes rapidly fade when the rotation stops because of retinal adaptation (15). The current view is that this phenomenon is related to the radial distribution of dichroic pigment molecules in the macula (16). We speculate that cross-polarized lenses may attenuate the PPR by inducing binocularly disparate patterns of retinal activation that result in altered patterns of cortical activation in the early but not, ordinarily, later stages of visual processing, and that these alterations render the stimulus less epileptogenic.
To identify patients who may benefit from blue cross-polarized glasses, in our laboratory, all patients with a history of PSE are now tested with photic stimulation at 3 to 30 Hz at a distance of 30 cm with the eyes closed. If this does not provoke PPRs, testing is repeated with the strobe at a distance of 15 cm, with the eyes closed, and then open. Patients are then tested while viewing a television screen at distances of 2 m and 30 cm. They are asked to bring computer games, hand-held electronic games, or videos that have induced seizures in the past. Stimuli are withdrawn as soon as PPRs are elicited and withheld for ≥20 s after the abnormal discharges have ceased. If PPRs are provoked in any of these scenarios, testing is repeated by using gray and blue nonpolarized, parallel-polarized, and cross-polarized glasses with similar overall transmittances.
Acknowledgment: Abstract presented at the Annual Meeting of the American Association of Electrodiagnostic Medicine in San Francisco, CA, USA, Sept 17-20, 2003