Presumed topiramate-induced retinopathy in a 58-year-old woman

Summary

We present a case of presumed topiramate-induced retinopathy in a 58-year-old woman who presented with progressive, bilateral visual loss following a 3- to 4-year history of oral topiramate intake for migraine. She reported difficulty with light adaptation, hemeralopia, and color desaturation. Her best-corrected visual acuity was 1/60 (20/1200) in the right eye and 6/18 (20/60) in the left eye, and she performed poorly on Ishihara color plate testing. Anterior segment examination was normal; dilated funduscopy showed mild macular pigmentary changes. Optical coherence tomography revealed subtle thinning and reduced reflectivity of the subfoveal ellipsoid zone and interdigitation zone bilaterally, associated with increased foveal autofluorescence. Humphrey visual field 24-2 revealed central defects. Electrodiagnostic testing showed a reduced and delayed b-wave and a normal a-wave on photopic full-field electroretinogram (ERG), with normal scotopic responses; multifocal ERG revealed reduced responses in the inner 10° in both eyes. She underwent extensive investigations including whole-body computed tomography and positron emission tomography scan, magnetic resonance imaging of the brain, uveitis screening, retinal autoantibody testing, and genetic testing on the retinal dystrophy panel to rule-out other causes for her presentation, all of which were normal or negative.

Introduction

Topiramate is an oral medication approved for treatment of epilepsy and for migraine prophylaxis.1 It has been shown to work by the following mechanisms: (1) enhanced GABA-mediated chloride fluxes across the post-synaptic membrane2; (2) positive modulation of the activity of GABA-A receptors3; kainite inhibition to activate the kainite/AMPA subtype of the excitatory amino acid receptor4; (4) carbonic anhydrase inhibition; and (5) state-dependent sodium channel–blocking action.5 Adult doses of topiramate are usually initiated at 25-50 mg per day; this is increased up to 50–100 mg a day for migraine prophylaxis and usually to 100–200 mg daily for epilepsy treatment.1 The effects of topiramate on the anterior segment of the eye have been well described. Richa and Yazbek6 noted that topiramate can cause acute, bilateral myopia related to anterior rotation of the ciliary body and suprachoroidal effusions. This can result in bilateral secondary acute angle closure glaucoma.7

The effects of topiramate on the retina are less well understood. Some studies have shown that 40% of retinal cells are immunoreactive to GABA, including subpopulations of amacrine and horizontal cells, bipolar cells, inter-plexiform cells, Müller cells, and retinal ganglion cells.8 A handful of cases of visual loss related to the effects of topiramate on the retina, but with different retinal features, have been reported previously.9–14 We present a unique case of presumed topiramate-induced retinopathy associated with severe, bilateral, irreversible visual loss, an electronegative cone electroretinogram (ERG) and abnormal multifocal ERG (mfERG), reduced color vision, subtle macular changes, and central visual field loss, which might represent the full spectrum of topiramate-related retinal toxicity.

Case Report

A 58-year-old woman initially presented with progressive reduction of vision bilaterally, with the right worse than the left eye. This occurred over a number of months and was described as a central blurring of vision that seemed to fluctuate. Peripheral vision was not affected. She was noted to have difficulty with light adaptation and hemeralopia and symptoms of color desaturation. She had no past ocular history. She had a past medical history of nonocular sarcoidosis, which had been treated many years previously with a course of oral steroids, hypertension, and recurrent migraines. Her medications included topiramate and oestradiol hormone replacement treatment. She had been taking topiramate for 3–4 years at a dose of 50 mg twice a day for migraine symptoms. There was no family history of inherited retinal diseases. She was a nonsmoker and drank <14 units of alcohol per week.

On initial presentation at Royal Liverpool University Hospital, St Paul’s Eye Unit, her best-corrected visual acuity was 1/60 (20/1200) in the right eye and 6/18 (20/60) in the left eye. She had reduced color vision bilaterally, with an Ishihara testing revealing correct identification of 1/17 plates in the right eye and 2/17 plates in the left, with the control plate identified bilaterally. There was no relative afferent pupillary defect on clinical examination. Anterior segment examination was normal, with deep anterior chambers and normal intraocular pressure in each eye. Dilated fundus examination revealed bilateral subtle central macular pigmentary changes and mild pigmentary changes in the inferonasal quadrant in both eyes (Figure 1) and healthy optic discs. No signs of intraocular inflammation were noted. Wide-field fundus autofluorescence images showed symmetrical mottled hypo-autofluorescence in the inferonasal quadrant in both eyes. Thirty-degree fundus autofluorescence images showed unmasking of the central foveal hypo-autofluorescence (Figure 2).

Figure 1.

Optos widefield color fundus photographs of the right and left eyes (A-B) showing subtle pigmentary changes in the inferonasal quadrant in both eyes associated with mottled hypo-autofluorescence (C-D).

Figure 2.

Thirty-degree fundus autofluorescence images showing unmasking of the central foveal hypoautofluorescence (right eye greater than left).

Optical coherence tomography (OCT) showed reduced reflectivity and thinning of the interdigitation zone (IZ) and reduced reflectivity of the ellipsoid zone (EZ) subfoveally in both eyes (Figure 3). OCT scans of the retinal nerve fiber layer were normal in both eyes. No fundus fluorescein angiograms were obtained, because there were no signs to suggest retinal ischemia; a macular OCT angiogram was normal, with no irregularities at the level of the different retinal vascular plexi or enlargement of the foveal avascular zone. Visual field testing (Figure 4) revealed bilateral central scotomata within the central 5° on 24-2 Humphrey visual fields, with peripheral areas of reduced sensitivity. Electrodiagnostic testing (Figure 5) showed normal scotopic responses but reduced and delayed b-waves on photopic full-field ERG (ffERG); mfERG revealed reduced responses in the inner 10° in both eyes, likely contributed by loss of cone function. The 30 Hz flicker ERG showed reduced amplitude and longer latency, indicating cone dysfunction.

Figure 3.

Macular OCT scans showed subtle thinning and reduced reflectivity of the interdigitation and ellipsoid zones subfoveally in both eyes (A, right eye; B, left eye).

Figure 4.

Humphrey Visual fields 24-2 testing showing bilateral central scotoma within the central 5° and additional areas of reduced sensitivity in the periphery.

Figure 5.

Electrodiagnostic testing. On full-field electroretinograms (ERGs), the scotopic response (top row) was normal, but there was reduced and delayed b-wave and a normal a-wave on photopic ERG (middle row). Multifocal electroretinogram (ERG) (bottom row) revealed reduced responses in the inner 10° in both eyes (right eye, left images; left eye, right images).

Given her symptoms, an electronegative ERG and absence of any striking findings on funduscopy, the patient was initially investigated to exclude cancer-associated retinopathy (CAR) with whole-body computed tomography and positron emission tomography and magnetic resonance imaging of the head, which returned no significant findings.

Other investigations, including a uveitis screen and retinal autoantibodies against recoverin and alpha-enolase, were negative; genetic testing on the retinal disorders panel by next-generation sequencing revealed no evidence of inherited retinal disease. Sarcoid uveitis was excluded with lack of ocular inflammation and a negative uveitis screen. Blood workup, including serum angiotensin converting enzyme level at 42 U/L, was normal. Following these extensive investigations, which ruled out other possible causes, presumed topiramate retinopathy was diagnosed We advised the cessation of topiramate, at this time patient’s visual acuity had decreased to 6/60 (20/200) from 6/18 (20/60) in the left eye.

Electrodiagnostic testsrepeated 8 months after the patient discontinued topiramate showed similar abnormalities without sign of recovery. At her most recent follow-up, 24 months after initial presentation, there has been no improvement in visual acuity, which has stabilized at 2/60 (20/600) in the right eye and has fluctuated between 4/60 (20/300) and 6/60 (20/200) in the left eye without overall deterioration. Her fundal changes and OCT features have remained stable, with no worsening or resolution. Visual fields were not repeated, because vision has remained too poor for reliable automated visual field testing for comparison with previous fields. She has since been commenced on sumatriptan for migraine prophylaxis.

Discussion

We conducted a literature review of ocular side-effects of topiramate focusing on retinal adverse effects. Yeung et al9 reported a case of a 48-year-old woman who developed blurred vision after 9 months on topiramate. The patient developed visual field constriction and fundus examination revealed bilateral diffuse pigmentary retinopathy, with reduced peripheral autofluorescence. There was no family history of retinal disease. Visual function improved after cessation of topiramate, which was believed to be the most likely cause of her retinopathy. In this case, ERGs remained unaffected.

Tsui et al15 reported a patient with decreased visual acuity who had been taking topiramate for 4 years. She was found to have bilateral vitelliform maculopathy on funduscopy. Her electro-oculogram showed normal Arden ratios, but she was found to have an electronegative ERG, with reduced b-wave amplitude in both eyes, which was not explained by vitelliform macular dystrophy.16 The authors concluded that the cause of the reduced b-wave amplitude could be topiramate use. This patient was not genetically tested to rule-out retinal dystrophy.

Vaphiades et al10 reported the case of a 32-year-old woman who complained of mildly reduced visual acuity 8 weeks after starting topiramate. Examination showed a central scotoma and reduced color vision; mfERGs showed central depressed waveforms. Although the patient was not extensively investigated, the temporal relationship made topiramate-induced maculopathy highly likely. No full-field ERGs were reported in this case.

Another case of presumed topiramate-induced maculopathy was reported by Beyenburg et al12 in a 41-year-old woman who developed maculopathy 3 years after starting topiramate. Examination showed mild reduction in vision and a central scotoma on visual fields testing. Her color vision was normal. No electrodiagnostic tests were reported.

Topiramate has also been associated with macular striae. In both reported cases,13,14 the striae disappeared shortly after topiramate cessation; no electrophysiological studies were performed. Gualtieri et al14 reported no peripheral choroidal effusions; normal anterior chamber depth was noted in their patient.

While these cases present a host of findings attributable to topiramate-induced retinopathy, what was different in our patient was that there were both macular and peripheral retinal changes and that her vision was more severely affected. Our patient also did not regain her vision after cessation of topiramate, in contrast to other reported cases. One unique finding in our patient was the disruption to the normal appearance of the subfoveal EZ and IZ on macular OCT, which could explain the permanent loss of vision. ERGs seem to show attenuation with prolonged exposure to topiramate, as in our case and as reported by Tsui et al.15

Electrodiagnostic tests have been undertaken three times in our patient during 2 years’ follow-up, and the fact that there has been no progressive deterioration in ERG findings has meant that the suspicion of CAR is low, although it is known that EDT findings can precede the onset of malignancy; our patient has undergone extensive testing to rule out any present concerns of occult malignancy. Although not all autoantibodies associated with CAR are known, and testing for all known autoantibodies has limitations, the most common antibody associated with CAR, namely anti-recoverin antibody, as well as the anti-enolase antibody, which is typically associated with cone dysfunction, were both negative in our patient.

Nutritional deficiencies that can cause, for example, bilateral central scotoma, reduced vision, and dyschromatopsia were not concerning, because our patient had no relevant history (eg, alcohol abuse, smoking), had a good diet, and examination and subsequent workup did not suggest optic neuropathy.

The mechanism of topiramate-induced retinopathy is unknown; we can attempt to understand the same by reviewing mechanisms implicated in the case of medications with a similar mode of action.

Cases of retinopathy have been reported with vigabatrin use which is also a GABA-enhancing medication.17 Krauss et al 18 reported a case series of patients using vigabatrin with visual loss wherein ERGs showed a selective reduction in b-wave activity, particularly in the cone pathway. Miller et al19 conducted electrodiagnostic studies of 39 patients on vigabatrin and found changes in b-wave ERG as well as reduction in oscillatory potentials. The authors postulated that the electrophysiological changes seen were due to GABA accumulation and activation of GABA receptors, in particular GABA_a, which is thought to reduce b-wave amplitude. In order to evaluate the effects on cessation of the medication, Johnson et al20 studied changes in patients with abnormal ERG associated with vigabatrin after their medication was discontinued for at least 3 months. In this case series, although there were reports of improvement of visual function in some individuals, there was no overall change in visual function or improvement in ERG as a group. Although vigabatrin has a half-life of 3 days, the study concluded that the medication has a permanent effect on retinal physiology. Thus, selective affliction of photopic b-wave is known with GABA-enhancing drugs and can explain cone ERG changes in our patient. There are established reports of retinopathy from vigabatrin use but very few case reports relating retinopathy to topiramate, which has a similar mechanism of action.

Kjellstrom et al21 conducted an animal study to evaluate the electrophysiological effects and histological changes to the retina with topiramate use. Initially, before treatment the ffERG results were similar in the 6 treated rabbits and in the 6 controls. After 8 months of treatment, light-adapted 30 Hz flicker was significantly reduced in the treatment group compared with the control group and baseline measurement prior to treatment. ffERG b-wave amplitudes from all treated rabbits also showed reduction in response to white light flash compared with baseline. Histological and immunohistological evaluation of the retina demonstrated increased GABA immunoreactivity in the amacrine cell bodies as well as the inner nuclear layer (INL), with loss of architecture in the inner plexiform layer as compared to the control. This study concluded that topiramate can cause characteristic abnormal ERG findings in rabbits and demonstrated links to GABA accumulation in amacrine cells and the INL. Reduced b-wave with GABA agonism has been described in other animal studies22; however, there are no studies conducted on long-term effects of GABA agonism on amacrine cells. Drawing from knowledge of idiopathic macular telangiectasia, in which loss of Muller cells, whose cell bodies are in the INL, leads to unmasking of central foveal autofluorescence and GABAergic pathways are implicated,23 its effects on the INL may explain the unmasking of central foveal autofluorescence in our patient. Although the clinical features and collaborative investigative results can be explained by the discussed pathogenic mechanisms related to topiramate-induced retinotoxicity, with this being an isolated case, causality is difficult to establish definitively.

In conclusion, we present a case of presumed topiramate retinopathy with bilateral severe irreversible central visual loss, macular and peripheral subtle pigmentary retinopathy, disrupted foveal anatomy on retinal imaging particularly obvious as persistent altered ellipsoid and interdigitation zones on macular OCT, characteristic b-wave ERG attenuation from selectively impaired cone pathway—all of which together appears to be a fuller manifestation of topiramate-induced retinopathy than that described previously in the literature. The study in rabbits provides evidence that topiramate use alters retinal electrophysiology and is associated with accumulation of GABA in the amacrine cells and the inner nuclear layer. Given the rarity of case reports relating to topiramate retinotoxicity, our case further highlights a potential rare side effect of topiramate. Patients should be counselled appropriately prior to commencing treatment with topiramate, and both prescribing clinicians and ophthalmologists should be aware of the features of topiramate-induced retinal toxicity. Early diagnosis is critical as some cases in the literature suggest that early cessation may mitigate the risk of severe and lasting visual loss.