We thank Qiu and associates for their comments relating to our publication, “Retinal Nerve Fiber Layer Thickness in Patients Receiving Chronic Anti–Vascular Endothelial Growth Factor Therapy.” Given their insightful concerns regarding the optical coherence tomography (OCT) inclusion parameters used in our study, we will take this opportunity to add additional information that may be helpful. In the methods section of our manuscript we state, “All scans were evaluated to ensure that the RNFL was accurately identified and that the signal strength scores were all equal to or above 6.” Further review of our data reveals that all scans had a signal strength (SS) ≥7. Furthermore, separate scans used for analysis for each patient did not vary in SS by more than 2 units. We believe this additional information illustrates why differences in SS between scans used in our study would be unlikely to lead to significant differences in retinal nerve fiber layer (RNFL) measurements.
It has been suggested that there is a positive correlation between SS and RNFL thickness. A study by Vizzeri and associates describes that for each unit decrease in SS, there is approximately a 2-μm decrease in RNFL thickness. The authors suggested that this may be a reason for test-retest variability with Stratus OCT. They concluded that a decrease in SS by 4 units or more from baseline with an associated decrease in RNFL could be mistaken for progression. Wu and associates studied OCT scans from 165 patients with glaucoma or optic atrophy and found no statistical differences in RNFL measurements with SS of 7 or more. A study by Samarawickrama and associates in adolescents found no significant effect of a decrease in SS on RNFL measurements. By excluding images with SS less than 6, as done in our study, the potential for image quality to significantly influence reproducibility of RNFL measurements is minimized. We agree there can be significant variability in RNFL measurements if scans are not centered on the optic nerve. While we reviewed all images in our study to ensure proper circle centration, newer OCT machines with tracking capabilities may provide additional information that can add to data presented in our study.
Finally, Qiu and associates state, “It may be a bit difficult to obtain good quality images in these patients who always presented with poor visual acuity.” As can be surmised from our published data, most of the patients had unilateral exudative disease and a subsequent review of the records reveals that the contralateral eye retained central vision. This is an excellent situation where the external fixation light is useful so that the nonimaged eye can be used for focusing and alignment of the scan for the contralateral eye undergoing imaging. All OCT studies obtained in clinical practice should be reviewed for quality, as indicated by SS, and for scan circle centration on the optic nerve. It is also imperative to know how to properly evaluate these parameters when reading published literature and to comprehend their limitations in influencing our clinical decision making.