advertisement
Lee et al. (187) should be commended for their well-designed study on evaluating the ability of time-domain optical coherence tomography (StratusOCT) to detect focal retinal nerve fiber layer (RNFL) defect progression. The investigators first determined the reproducibility of global and sectoral RNFL measurements obtained by the device. Based on test-retest variability, OCT progression was defined as eyes that exceeded the 95% confidence interval level of this parameter. Using these cut-offs they tested the ability of the device to detect RNFL progression in eyes that were subjectively judged to progress by red-free photography. The authors reported a specificity of 59.7% and sensitivity of 85.2% when progression was noted in at least one clock hour and 98.4% and 14.8% specificity and sensitivity when judging by the overall RNFL thickness. Repetition of the OCT test improved the specificity of the clock-hour analysis to 92.5%. Because the authors selected subjects with focal RNFL abnormalities it is not surprising that the localized OCT analysis detected more abnormalities than the global analysis, as reflected by these results. The authors also reported a good spatial relationship between the location of progression as defined by photography assessment and OCT measurements. While the study demonstrated the ability of OCT to detect RNFL defect progression, the clinical utility is currently limited because the commercial device does not provide this type of information. Moreover, the studied population included subjects with clearly visible RNFL defect progression in red-free photography ‐ encompassing only part of glaucoma patients in a typical clinical setting.