advertisement
Imaging instrumentation, producing measurements of the optic disc, neuroretinal rim, and retinal nerve fiber layer, was introduced in the early 1990s with refinements occurring at approximately five-year intervals. As manufacturers offer new devices, the challenge for clinicians is to understand whether these instruments are an improvement, and how to incorporate them into practice. The authors address these questions. Ideally, one would want a new instrument to be backwards compatible so that data from the old instrument could be used with new data for management purposes. Unfortunately, this is not always the case. Also, one measure to understand how well a new instrument performs is in its test-retest ability.
Spectral-domain (SD) optical coherence tomography (OCT) was recently introduced, replacing the time-domain (TD) variety. SD OCT offers distinct advantages, such as better scan resolution due to its quicker speed with up to nine different manufacturers building devices compared to one for TD-OCT. Unfortunately, none of the new instruments are backwards compatible with the old one. In this study, Vizzeri et al. (555) evaluate TD and SD OCTs from the same manufacturer (Carl Zeiss Meditec). One objective was to assess SD OCT reproducibility (test-retest) and the other was to compare measurements taken on the same day between the instruments. With regard to SD OCT reproducibility, the measurements were impressive with CV and ICC for average RNFL thickness being from 1.6% to 0.98% in patient eyes. The measurements between the TD and SD OCTs were found to correlate, but were not similar, with TD measurements being consistently thicker indicating that one needs to baseline each person when switching to a new instrument. Measurements taken on one instrument cannot be used or compared to those taken from a different instrument. It also shows the importance of a normative database, as this will be the best guide when a clinician is analyzing the results.