Dev Med Child Neurol. 2015 Jan;57(1):60-7.
Does vigabatrin treatment for infantile spasms cause visual field defects? An international multicentre study.
Riikonen Sylvia Raili1, Rener Primec Zvonka2, Carmant Lionel3, Dorofeeva Maria4, Hollody Katalin5, Szabo Ilona 6, Stirn Krajnc Branka7, Wohlrab Gabriele8, & Sorri Iiris9
1Children`s Hospital, University of Eastern Finland, Kuopio, Finland
2 Department of Child, Adolescent and Developmental Neurology, University Children`s Hospital, Ljubljana, Slovenia
3Department of Pediatrics, University of Montreal, Quebec, Canada
4Department of Neurology & Epileptology, State Institute of Pediatrics& Pediatric Surgery, Moscow, Russia
5Department of Ophthalmology, University of Pecs, Pecs, Hungary
6Department of Ophthalmology, University Medical Centre Ljubljana, Slovenia.
7 Department of Ophthalmology, University of Medical Centre Ljubljana
8University Children`s Hospital, Zurich, Switzerland
9Department of Ophthalmology, University of Eastern Finland, Kuopio, Finland
Objective The present study is the largest study reporting the results of visual field testing in children who started VGB treatment during the first year of life.
Method Thirty-five children were studied at school age to see whether VGB treatment in infancy had caused visual field defects (VFDs).
Visual fields were examined by the Goldmann kinetic perimetry, by the static Humphrey or Octopus perimetry by perimetrist. Visual fields were re-evaluated by the ophthalmologist.
Results Typical VGB-attributed visual field defects were found altogether 11/32 (34%) patients. The defects were mild in five and severe in 6 cases. One child out of 12 children (8%) who used VGB for less than one year (Group 1) had mild VFD. Three of 9 patients (33%) using VGB for 22-24 months (Group 2) and 7/11 patients (63%) using VGB for more than 2 years (Group 3) had VFDs. Defects were mild (1) Group1 and mild (1) and severe (2) in Group 2), and mild (3) and severe (4) in Group 3. The mean cumulative doses of VGB were 140, 5 g (Group 1), 758,8 g (Group 2) and 2712g (Group 3), respectively. Patients with tuberous sclerosis had more VFDs (6/10 patients). The majority of children in our study were treated over a long period and with large cumulative doses of VGB. VFDs were found in 34 % of the cohort. The rate of VFD increased from 9% to 63 % as duration of treatment increased.
Interpretation VFDs were found to occur at the same frequency as in adults. The risk/benefit ratio should always be considered when using VGB.
VGB has gained popularity in the treatment of IS. It is only licensed as first-line /monotherapy for infantile for the treatment of IS.
In our study (Ref 1) the visual field and ophthalmological findings were assessed in each center (8) and re-evaluated by an experienced ophthalmologist. VGB-attributed VFDs were considered as a bilateral, symmetrical nasal, or concentric constriction of the peripheral visual fields, as characterized in several studies. Normal visual fields were defined as a temporal meridian extending beyond 70° and nasal meridian extending beyond 50°. Mild VFDs were defined as in previous studies, that is temporal meridians lying between 50° and 70°, and severe VFDs as meridians as <50°. Additional requirements for normality were the superior field extending to 40° and inferior to 50°. Abnormal findings were confirmed by repeat testing. From our study it can be concluded that VGB treatment for infantile spasms can cause visual field defects at the same frequency as that reported in adults. Short treatment duration correlates with lower incidence of VFD. Duration of VGB treatment less than 1 year seems to be safe. The plasticity of an infant retina does not to protect from damage.
VGB is a selective irreversible inhibitor of GABA-transaminase.
Several theories have been proposed about the cause of VGB-associated VFDs. These include toxic reaction (treatment duration and cumulative dosage, daily dose, idiosyncratic reaction, polypharmacy, pharmacogenetic difference, elevated levels of retinal ornithine, light exposure, taurine deficiency, GABA accumulation in the retina, dysfunction in GABA-innervated horizontal and amacrine cells in the inner plexiform layer of the retina, and the infantile spasm itself. However, the most viable hypothesis to date, is that not GABA itself that affects the cones of retina, but rather VGB.
In elder children and adults VGB causes irreversible visual field defects in about a third of patients (34-40 %, respectively). In a recent study from UCLA 2016 there was a lack of clinically apparent vision loss in a large number of patients with IS studied but the study was done without aid of perimetry or serial ERG and patients were young. Young children are not able to cooperate with visual field testing. Furthermore, the patients had a short treatment duration.
ERG has been studied as possible method to detect the toxic effects of VGB on the retina during infancy while the drug is still been used. Shortest time for ERG abnormalities reported is 3 months. Dragas et al (2014) (REF 2) have tested 100 children with IS treated with VGB (median age at baseline: 7.6 months; range 1.7-38.4) for the long flash ERG). Normal age-matched controls were also tested. The cone d-wave amplitude in patients VGB was associated with retinal defects. Longer duration of treatment was associated with reduced d-wave amplitude (ANOVA p ≤ 0.05). Their inclusion criteria required a reduction in ERG amplitude from baseline to exceed 37% and be sustained over at least 2 subsequent testing session. The prevalence of VGB-attributed defect was comparable to our study. However, it remains unclear whether ERG is useful for VGB monitoring during infancy. Examination of the retinal nerve fiber layer using optical coherence tomography (OCT) has recently been introduced as an alternative method for follow-up of VGB toxicity. However, optical coherence tomography also requires the cooperation of the patient. Regularly performed visual field examination remains the cornerstone in screening.
In 2009, VGB was approved by US Food and Drug Administration approved as monotherapy for paediatric patients for whom the potential benefits outweigh the risk of vision loss. Testing (Lundbeck Pharmaceuticals programme), should be done at baseline and at least every 3 months while the patient is taking VGB. This requirement would, however, be extremely difficult to implement, as visual field examination cannot be carried out in young infants using current methods. There is required by the FDA that patients who do not obtain the scheduled ophthalmic testing be tapered from VGB.
Hormonal treatment was preferred over VGB in one evidence-based review and in one recent multi-centric study from US. It was found that it controls spasms sooner and more frequently than VGB does. Hormonal treatment may also result in better long-term outcomes in cryptogenic cases. Because of the severe side effects of both drugs, doses should be low and treatment periods short.
Figure 1. The girl with cryptogenic spams showed severe VFD (right eye). She had used VGB for 22 months with highest daily dose of 750 mg and cumulative dose of 495 g.
Riikonen R. Rener-Primec Z, Carmant L, Dorofeeva M, Hollody K, Szabo I, Kranjnc B, Wohlrab G, Sorri I. Does vigabatrin treatment for infantile spasms cause visual field defects? An international multicentre study. Dev Med Child Neurol 2015. See comment in PubMed Commons belowJan;57(1):60-7. doi: 10.1111/dmcn.12573. Epub 2014 Aug 22.
Dragas R, Westall C, Wright T. Changes in the ERG d-wave with vigabatrin treatment in pediatric cohort. Doc Ohtalmol 2014; 129: 97-104. doi: 1007/s10633-014-9453-y.
Dr. Riikonen Sylvia Raili