Statistics from Altmetric.com
We discussed the article by Cobb et al1 with great interest in our journal club meeting. We would like to highlight some of the issues we discussed.
Evidence in the literature2,3 suggests anisometropia is a difference of 1 DS between both eyes but the authors use a difference of 2 DS. Also the minimum criteria for diagnosing amblyopia on a test of visual acuity are accepted to be two lines difference between the eyes on linear tests. However, the authors use these criteria for the single optotype test and do not take into account the crowding effect. A two line difference on Sheridan Gardiner may actually equate to a three line difference on a linear test.2 The Royal College of Ophthalmologists3 also state that on repeated testing a difference of one line can also be included—that is, 6/9 6/6. Therefore a one line difference on Sheridan Gardiner may actually fulfil the linear criteria.
The authors stated that part time occlusion was the method of treatment. Part time occlusion can vary from as much as 6 hours to 10 minutes yet the authors did not state the amount of occlusion to the level of vision. Methods of occlusion have varied greatly between 1972 and 1995, so were the guidelines for occlusion treatment the same throughout the study period?
Furthermore ,”compliance” with treatment is recognised as an important factor affecting the final visual outcome in amblyopia3 and that compliance varies with age.4 The authors have not analysed if compliance varied significantly with age of presentation.
It has been shown that anisometropic amblyopes can have improvement in vision after a period of spectacle correction and refractive adaptation is considered an important component of amblyopia treatment.5 The authors did not state if occlusion was started at the first, second, or third visit after spectacle correction. We do not know if treatment had been initiated prematurely and resulted in falsely increased occlusion dose, thereby giving poorer compliance. This could have skewed the figures. So, we recommend exercising caution in interpreting the results and in concluding that time at which screening is carried out, not critical in this group.
We thank Sankari and colleagues for their helpful and interesting comments regarding our paper. We accept that the literature suggests a difference of 1 D as a definition of anisometropia but felt the stringent definition we applied would more accurately represent a degree of anisometropia that could potentially lead to amblyopia, and therefore makes the findings more robust. The average age of the cohort was 5½12 years and as such the majority of tests were linear. Pure anisometropic amblyopia does not tend to cause as significant crowding as strabismic. We therefore felt that it was not necessary to equate one line of difference with Sheridan Gardiner to a two or three line difference on a linear test. Occlusion regimes and compliance are always problematic especially in retrospective studies.1 All parents of children were given the same advice and support regarding therapy and the outcomes, therefore the findings of our study reflect a real clinical situation. Compliance does vary with age as shown by Nucci et al,2 but this was significant only in the 1–2 year old age group. There were no children of this age in our study. More importantly, the effect of occlusion is inversely related to age, so the younger the child the more rapid the response. Most children were commenced on occlusion at their first visit after refractive correction, which was within 2 months of receiving glasses. We agree that anisometropic amblyopes may have a continuing improvement in their vision with just spectacle correction for some months.3 Depending on the response to spectacle correction, this department will now delay occlusion for up to 4 months. To suggest that initiating occlusion earlier than may have been necessary would result in a worse outcome because of poor compliance must be considered speculative. In any case, failure to comply with occlusion would result in the now recommended treatment—that of spectacle correction alone. While there may be some debate as to the efficacy of occlusion therapy,4 there is currently no evidence that it is detrimental to the final visual acuity achieved in anisometropes.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.