advertisement
Ortega et al. (186) aimed at evaluating the frequency and the type of imaging artifacts in the optic nerve head analysis protocol of Stratus OCT in glaucomatous and normal eyes. The authors found a 12% occurrence of artifacts over a retrospective series of 264 (146 normals and 118 with glaucoma) eyes. The study is single-center, retrospective and results may be scarcely generalizable, strongly depending on the quality of image acquisition and experience of the grader. Actually, the authors used the stringent selection criteria that are usually adopted for OCT studies. It should have been pointed out that in clinical practice, the rate of artifacts is probably higher than shown in the paper (high refractive errors, cataract etc) and often the presence of artifacts may not be so easy to demonstrate (in the paper agreement required a third grader in 10% of artifacts). Most of artifacts were manually (and very empirically) corrected by modification of retinal surface sensitivity, though there were cases that could not be changed at all. The impact of such changes on the 'performances' of ONH protocol for follow-up are not known and should be evaluated. The key message ‐ that is that all scans should be carefully checked for quality in order to avoid, as much as possible, the presence of artifacts ‐ seems useful, though quite obvious. Overall, the paper confirms that high-tech instruments are helpful for correct diagnosis and for glaucoma follow-up, yet they cannot be a proper surrogate of the clinical examination of the eye with glaucoma. The findings of this study are interesting as they stress the fact that protocols should be continuously improved and checked. Perhaps more discussion should have been dedicated to reliability and interpretation of high-tech instruments for glaucoma in general, and OCT in particular.