We thank Drs Gonul and Okudan for their interest in our study. In their letter, they mention the inaccuracy of manually performed measurements of choroidal thickness and suggest to average several manual measurements taken at different locations in the same horizontal section of the macula. One may argue against this suggestion, since the choroidal thickness markedly gets thinner the more distant from the foveola. Interestingly, we found that the eyes with acute nonarteritic anterior ischemic optic neuropathy had thinner choroidal thickness measurements in the foveola and in all other measurement locations except for the nasal location. We agree with Drs Gonul and Okudan that an automatic delineation of the choroidal borders may be helpful for choroidal thickness measurements.
Drs Gonul and Okudan state, and we fully agree, that choroidal thickness depends on a variety of ocular and systemic parameters such as age, spherical equivalent, and axial length. Since we did not adjust for all of these parameters in our study, Drs Gonul and Okudan discussed that the thin subfoveal choroid detected in the patients of our study might have been attributable to underlying systemic diseases and might not have primarily been associated with the acute nonarteritic anterior ischemic optic neuropathy. We would like to bring forward the argument that in large population-based studies such as the Beijing Eye Study, systemic diseases including arterial hypertension and diabetes mellitus did not have a major influence on subfoveal choroidal thickness. Since the choroidal thickness measurements had been adjusted for the main factors (ie, age and refractive error) influencing choroidal thickness, one may infer that the marked difference in subfoveal choroidal thickness between patients with acute nonischemic optic neuropathy and normal individuals in our study may not have been markedly influenced by differences between both groups in the prevalence of underlying systemic diseases.
Drs Gonul and Okudan cited a previous study by Dias-Santos and associates in which subfoveal choroidal thickness was significantly thicker in eyes at 57.2 ± 26.9 months after an acute nonischemic anterior ischemic optic neuropathy than in the control group. This appears to be in contrast to our investigation. In the study by Dias-Santos, however, subfoveal choroidal thickness increased with longer follow-up after the event ( P = .047; correlation coefficient r = 0.46). Taking into account the mean follow-up and extrapolating the regression line of the association between subfoveal choroidal thickness and follow-up time, one arrives at a subfoveal choroidal thickness of about 190 μm at the time when the nonischemic anterior ischemic optic neuropathy occurred. That is approximately the same value we found in our study on patients with an acute nonischemic anterior ischemic optic neuropathy. Both values of subfoveal choroidal thickness, as determined by extrapolation in the study by Dias-Santos and as measured in our study, are thinner than the mean value of subfoveal choroidal thickness (254 ± 107 μm) measured in a population-based study on 3233 Chinese individuals after adjustment for age and refractive error. In investigations on individuals of other ethnicity, similar measurements of subfoveal choroidal thickness were found as in the Beijing Eye Study. In conclusion, the previous study by Dias-Santos and associates and our study both support that eyes with an acute nonischemic anterior ischemic optic neuropathy have a thinner subfoveal choroid as compared to normal eyes.