Are GABAergic Agents Toxic to the Retina?
Abstracts & Commentary
Sources: Krauss GL, et al. A controlled study comparing visual function in patients treated with vigabatrin and tiagabine. J Neurol Neurosurg Psychiatry. 2003;74:339-343; Lawden MC. Vigabatrin, tiagabine, and visual fields. J Neurol Neurosurg Psychiatry. 2003;74:286.
Vigabatrin (VGB) is an antiepileptic drug (AED) whose putative mechanism of action is to increase inhibitory tone in the brain by blocking the metabolism of gamma-amino butyric acid (GABA) by irreversibly binding to GABA transaminase. It is available in the European Union, Canada, and Mexico, where it is the AED of choice for treating infantile spasms associated with tuberous sclerosis. VGB is unlikely to achieve approval by the US Food and Drug Administration because up to 40% of patients treated with the drug develop concentric visual deficits, which can be permanent. While there have been case reports of tiagabine (TGB), an AED that blocks GABA reuptake by neurons and glia, causing comparable visual impairments, Krauss and colleagues are the first to undertake a controlled study to compare VGB and TGB with regard to this adverse effect.
The study design involved a cross-sectional comparison of visual acuity, color vision, Goldman and Humphrey perimetry, and electroretinograms (the effects of VGB on vision are thought to be due to retinal dysfunction) among patients taking VGB, TGB, or another AED (the control group) for at least 6 months. Thirty-two patients were treated with VGB, 12 with TGB, and 14 with another AED. Eight of the control subjects were taking an agent known to affect GABA: gabapentin, topiramate, or valproic acid. The main findings were that there were no differences between groups when looking at visual acuity or color vision. However, the VGB group had abnormal findings on kinetic (Humphrey) visual field testing and ERG that were significantly different relative to TGB and control patients.
Commentary
There is extensive clinical experience with GABA agonists that bind to the barbiturate and benzodiazepine binding sites on different classes of the GABA receptor. There have been no consistent findings of visual dysfunction in patients thus treated. Rather than binding directly to GABA receptors, both VGB and TGB modulate GABAergic tone by modulating synaptic accumulation of GABA. Why, then, are there significant differences in the adverse effect profile of these 2 agents with respect to retinal function? Krauss et al cite several theories to account for this phenomenon. One hypothesis relates to the fact that VGB markedly increases the concentration of GABA in cortical astrocyte cell culture, whereas TGB does not.1-2 These supra-physiologic GABA levels may play a role in GABAergic amacrine cell injury seen in the inner retina in pathologic specimens. There are also differences in VGB distribution in the retina relative to brain. VGB levels are 5-fold higher in rodent retina vs brain. Conversely, TGB levels are slightly reduced in retina vs brain.3
Clearly, not all GABA agonists are the same. The lesson of VGB vs TGB is that not all modulators of GABA reuptake and metabolism are the same, either. The latter fact is promising with respect to the rational design of drugs that modulate GABA. As Lawden concludes in the accompanying editorial to the study: "A class effect of GABAergic drugs causing retinal damage now seems unlikely." These agents are quite effective in the treatment of epilepsy. We can look forward to new AEDs currently in development that increase GABAergic tone without the retinotoxic effects of VGB. — Andy C. Dean
Dr. Dean is Assistant Professor of Neurology and Neuroscience, Director of the Epilepsy Monitoring Unit, Department of Neurology, New York Presbyterian Hospital—Cornell Campus.
References
1. Sills GJ, et al. Seizure. 1999;8:404-411.
2. Fraser CM, et al. Epileptic Disord. 1999;1:153-157.
3. Sills GJ, et al. Neurology. 2001;57:196-200.