Anisometropia and amblyopia--chicken or egg?

Editorials Anisometropia and amblyopia-chicken or egg? Amblyopia continues to attract tremendous interest from basic scientists and clinicians, yet surprisingly large gaps in our understanding of this topic remain. One of these is highlighted by Abrahamsson and Sj6strand who describe in this issue of the B7O (p 860) a meticulous follow up of 20 children to the age of 10 who at 1 year of age had .3.0 dioptres of anisometropia. Three distinct patterns of development emerged. Firstly, in some the anisometropia decreased (mean 3.0 DS) and no amblyopia resulted. A second group all developed amblyopia and the aniso-metropia increased (mean 1.4 DS), while in a third group the anisometropia decreased to a lesser extent than in the first group (mean 1.2 DS) and all developed squint and/or amblyopia. At the outset, all infants studied had received full refractive correction so one could quibble with the authors' use of the term 'natural history' in referring to the development they observed. Nevertheless, their work draws attention to how little we know about the development of abnormal refractive states and, specifically, anisometropia. The question now posed is which occurs first, amblyopia or anisometropia? Hitherto, it has been generally assumed that anisometropia precedes amblyopia; indeed, this is said to differentiate the amblyopia of micro-tropia from that occurring in larger angle strabismus.' However, the direction of causality in this relation has been questioned.2 Six years ago Almeder and colleagues wrote '...despite the clear association of anisometropia with amblyopia and despite demonstrations from animal models that anisometropia can cause amblyopia, it has not been proven that the anisometropia seen in amblyopic school children or adults is the cause of that amblyopia." By 1992 an alternative opinion was forming '...we are beginning to suspect that anisometropia, at least in the population we studied, may be more often a result than a cause of amblyopia.'4 To most clinicians this is counter-intuitive yet this hypothesis in fact arose out of clinical practice. Lepard' and, later, Nastri and colleagues6 observed that in unilateral amblyopia the refractive status of the fixing eye becomes more myopic while the amblyopic eye remains hypermetropic. Refractive development studies in normal and strabismic children have shown that the incidence of anisometropia in infants can be as high as 25% and in early childhood is between 2.8% and 8%.' While the prevalence of anisometropia throughout childhood is relatively constant, the children who make up this group are …