Reply

We understand that many readers may have concerns about the conclusions drawn by our study. We were not aware of any teaching that bifocals are provided to young children for “comfort,” and according to the American Academy of Ophthalmology Preferred Practice Pattern for esotropia: “Bifocal treatment should be considered in patients with potential for sensory fusion…. If successful, bifocals may be necessary on a long-term basis to maintain binocular alignment for viewing near targets.”¹ In our clinical experience we have not encountered children who report increased comfort with bifocals when they are first prescribed. To the contrary there can be a long and difficult period of adaptation, which is one of the observations that inspired one co-author to discontinue the practice of prescribing bifocals. Certainly we do see children who have been treated with bifocals and who, as a result, must use their bifocal for near work; these patients have likely lost their accommodative ability and/or their fusional divergence and so have become dependent on bifocals. Indeed, von Noorden et. al.² reported that bifocal wear is associated with a “moderately or, in some cases, markedly reduced near point of accommodation.” We would be interested to see evidence that patients are more comfortable with bifocals, as such evidence would then need to be considered in the decision as to whether to prescribe in any one patient.

We also agree that there are a variety of potential known and unknown confounding factors that make it difficult to accept as clinically relevant the statistically significant, four-fold increase in the risk for surgery that we observed in patients treated with bifocals. We made every effort to adjust for any confounding factors, but statistical adjustments of retrospective data are inherently limited. Dr. Wright suggests that the bifocal group had a larger near deviation, but we both controlled for that, and conducted a subgroup analysis of patients with near deviation greater than 20 prism diopters (not shown), with the same results. He also suggests that the bifocal group might have had a higher AC/A ratio than the single vision group. Such a difference would have been entirely fortuitous considering the study design (where different clinicians prescribed or did not prescribe bifocals as a result of where and when they trained rather than on patient characteristics), but we are all in agreement that a prospective study is required to determine whether our observations should be translated to clinical practice. While we did not expect at the onset of our study that the bifocal group would have had worse outcomes, this is what we found. There remains no Level 1 evidence that bifocals are either helpful or harmful, and our results make it clear that the time has come to invest in a prospective, randomized, controlled trial to obtain a more definitive answer to this important clinical question.