Changes of choroidal structure after treatment for primary intraocular lymphoma: retrospective, observational case series.

BACKGROUND We report changes of choroidal structure determined by binarization of enhanced depth imaging optical coherence tomographic (EDI-OCT) images after treatment for primary intraocular lymphoma (PIOL). METHODS Five eyes of four patients with PIOL were examined by EDI-OCT before and 6 months after intravitreal methotrexate injections. In addition, 15 eyes of 15 normal individuals controlled by age and refractive error were examined by EDI-OCT. Binarization of the EDI-OCT images was performed using publicly accessible software (ImageJ). The examined area of the subfoveal choroid was 1,500 μm wide, and the dark areas that represented the luminal areas were traced by the Niblack method. Wilcoxon signed rank test was used to determine the significance of changes in the subfoveal choroidal thickness, interstitial area, and luminal area. Mann-Whitney U test was used to compare the parameters in the eyes with pretreatment PIOL and normal control eyes. RESULTS The subfoveal choroidal thickness was significantly decreased after treatment (P = 0.0431). In the binarized images, the interstitial area was significantly decreased after treatment (P = 0.0431), while the luminal area was not significantly changed (P = 0.8927). After delayed onset of PIOL, increased interstitial area, thickened choroid and unchanged luminal area were observed in one eye. The interstitial area and choroidal thickness were significantly increased in the eyes with pretreatment PIOL compared with the normal control eyes (P = 0.0207, P = 0.0495, respectively), while the luminal area was not significantly different (P = 0.2752). CONCLUSIONS After treatment for PIOL, the EDI-OCT images showed a thinner choroid, and binarization of the EDI-OCT images showed significantly decreased interstitial areas compared with the luminal areas. The binarized EDI-OCT images can provide useful information on choroidal structure in eyes with PIOL, and combining these images with intraocular interleukin levels or fundus autofluorescence images should provide valuable information for determining the PIOL activity.