Sir,

We thank Drs Salti and Khouri for their thoughtful comments. Certainly, digital imaging is taking an increasingly prominent role in the setting of retinal imaging and we expect that this approach will ultimately replace film imaging. However, for our study, we were interested in answering two specific questions that made film imaging preferable. First, as there is little published data on the evaluation of colour fundus photos for the identification and triage of neovascular AMD, we were interested in determining if this concept was viable. Hence, we desired to use ‘gold-standard’ imaging and not to introduce a second variable, that of digital imaging. Second, our goal was to evaluate both stereo and nonstereo images sets and, for this purpose, Kodachrome stereo imaging is well-established—digital stereo imaging techniques are still evolving. We also like to caution readers about the use of digital manipulation tools during the process of image interpretation. Depending on the algorithms used, data and potentially diagnostic accuracy, can be lost with contrast enhancement tools and sharpening filters.