Childhood blindness and visual loss: an assessment at two institutions including a "new" cause

This study was initiated to investigate the causes of childhood blindness and visual impairment in the United States. We also sought a particular etiology--congenital lymphocytic choriomeningitis virus (LCMV)--which has been considered exceedingly rare, in a fixed target population of children, the severely mentally retarded.

We undertook a library-based study of the world literature to shed light on the causes of childhood blindness internationally and to put our data in context. We prospectively examined all consented children (159) at 2 institutions in the United States to determine their ocular status and the etiology of any visual loss present. One of the institutions is a school for the visually impaired (hereafter referred to as Location V), in which most of the students have normal mentation. The other is a home for severely mentally retarded, nonambulatory children (hereafter referred to as Location M). This institution was selected specifically to provide a sample of visual loss associated with severe retardation because the handful of cases of LCMV in the literature have been associated with severe central nervous system insults. Histories were obtained from records on site, and all children received a complete cyclopleged ophthalmic examination at their institution performed by the author. Patients at Location M with chorioretinal scars consistent with intrauterine infection (a possible sign of LCMV) had separate consents for blood drawing. Sera was obtained and sent for standard TORCHS titers, toxoplasmosis titers (Jack S. Remington, MD, Palo Alto, Calif), and ELISA testing for LCMV (Centers for Disease Control and Prevention, Atlanta, Ga).

The diagnoses at Location V were varied and included retinopathy of prematurity (19.4%), optic atrophy (19.4%), retinitis pigmentosa (14.5%), optic nerve hypoplasia (12.9%), cataracts (8.1%), foveal hypoplasia (8.1%), persistent hyperplastic primary vitreous (4.8%), and microphthalmos (3.2%). The most common diagnosis at Location M was bilateral optic atrophy, which was found in 65% of the patients examined who had visual loss. Of these, the insults were most often congenital (42.6%), with birth trauma, prematurity, and genetics each responsible for about 15% of the optic atrophy. The second most common diagnosis was cortical visual impairment (24%), followed by chorioretinal scars (5%), which are strongly suggestive of intrauterine infection. Of 95 patients examined at Location M, 4 had chorioretinal scars. Two of these had dramatically elevated titers for LCMV, as did one of their mothers. One of the other 2 children died before serum could be drawn, and the fourth had negative titers for both TORCHS and LCMV.

At both locations studied, visual loss was most often due to congenital insults, whether genetic or simply prenatal. The visual loss at Location V was twice as likely as that at Location M to be caused by a genetic disorder. The genetic disorders at Location V were more often isolated eye diseases, while those among the severely retarded at Location M were more generalized genetic disorders. Our study identified optic atrophy as a common diagnosis among the severely mentally retarded with vision loss, a finding that is supported by previous studies in other countries. In our population of severely retarded children, the target etiology of lymphocytic choriomeningitis virus was responsible for half the visual loss secondary to chorioretinitis from intrauterine infection. This is more common than would be predicted by the few cases previously described in the literature, and strongly suggests that LCMV may be a more common cause of visual loss than previously appreciated. We believe that serology for LCMV should be part of the workup for congenital chorioretinitis, especially if the TORCHS titers are negative, and that perhaps the mnemonic should be revised to "TORCHS + L." Childhood blindness and visual impairment are tragic and co