advertisement
Editors Selection IGR 14-1
Clinical Examination Methods: How often should we scan the macula to catch progression?
Comment by Kouros Nouri-Mahdavi on:
127494 Number of macula optical coherence tomography scans needed to detect glaucoma progression, Mahmoudinezhad G; Moghimi S; Latif K et al., British Journal of Ophthalmology, 2025; 109: 675-681
The paper by Mahmoudinezhad et al. investigates the optimal frequency of macular optical coherence tomography (OCT) scans required to detect glaucoma progression in a timely manner. The investigators analyzed data from 332 eyes (glaucoma suspects or patients with mild to moderate glaucoma) from the Diagnostic Innovations in Glaucoma Study (DIGS) and the African Descent and Glaucoma Evaluation Study (ADAGES) with an average follow-up of 4.4 years utilizing computer simulations (200,000 sequences) to identify statistically significant changes in global ganglion cell complex (GCC) thickness measurements. This simulation assessed the time to progression at varying loss rates (-0.5 µm/year to -2.0 µm/year) and testing frequencies (one, two, or three times per year) with a statistical power of 80% to 90%. The findings indicate a significant enhancement in detection power when moving from one to two tests per year. However, the move from two to three tests yielded a much smaller gain. For a standard loss rate of -1.0 µm/year, the third test only accelerated detection by 0.4 years. Given these findings, semi-annual testing may optimize resource utilization and reduces healthcare burdens while maintaining the ability to identify clinically significant progression. Reimbursement issues for frequent OCT testing still need to be addressed. On the other hand, in research setting, it might be preferable to still use higher frequency testing (x3/year) as carrying out a trial is an expensive endeavor and the decrease in time to detect progression may be worth the cost and effort. One has to be mindful of the study’s limitations. The ETDRS subfields do not have the same size and hence the global measurements are not quite the same as the global GCC measurements calculated from macular superpixels or those provided by other OCT devices. It is not clear how this source of noise might have affected the outcomes. The improvement in detection also comes at the cost of a much higher false positive detection rate, 12% and 15% with twice yearly or x3/year testing frequency, respectively. This is an expected shortcoming of simple linear regression used to estimate GCC rates of change. Bayesian regression models, which are able to use information from the population, may be able to address this issue. Binning by glaucoma severity was appropriately done, but due to the nature of the cohorts used, eyes with severe glaucoma could not be separated from those with moderate stage glaucoma; therefore, the findings may not be generalizable to eyes with advanced glaucoma. Mahmoudinezhad and colleagues are to be commended for providing additional insight on this important topic in glaucoma diagnostics.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
