Purpose: To analyze retinal nerve fiber layer (RNFL) thickness in eyes with or without visual field (VF) defects after indocyanine green-assisted vitrectomy for idiopathic macular holes using optical coherence tomography (OCT) and to investigate the relationship between postoperative VF defects and RNFL damage. Design: Retrospective interventional case series. Participants: Thirty-four eyes of 32 patients with idiopathic macular holes that underwent vitrectomy between January 2001 and March 2003 were included in this study. Eyes were divided into 3 groups according to the occurrence of postoperative VF defects and the use of indocyanine green for internal limiting membrane peeling during surgery: 11 eyes with VF defects after indocyanine green-assisted vitrectomy (group 1),9 eyes without VF defects despite the use of indocyanine green (group 2),and 14 eyes without VF defects that underwent vitrectomy without indocyanine green (group 3). Methods: Retinal nerve fiber layer thickness in each of 4 quadrants (superior,inferior,nasal,temporal)was measured with OCT. Main OutcomeMeasure: Retinal nerve fiber layer thickness around the optic disc. Results: The mean RNFL thickness in 3 of 4 quadrants (superior,nasal,inferior) in group 1 was significantly less than that in the corresponding quadrant in groups 2 and 3 (P < 0.01). In the temporal quadrant,there was a significant difference between groups 1 and 3 (P=0.02),but not between groups 1 and 2. The RNFL thickness in group 1 was significantly less in 3 quadrants in operated eyes than in fellow eyes (P< 0.05). Conclusions: The RNFL thickness was reduced in eyes with VF defects after indocyanine green-assisted vitrectomy for macular holes,suggesting that the postoperativeVF defectsmay have been caused by RNFL damage relating to the use of indocyanine green. Purpose: To analyze retinal nerve fiber layer (RNFL) thickness in eyes with or without visual field (VF) defects after indocyanine green-assisted vitrectomy for idiopathic macular holes using optical coherence tomography (OCT) and to investigate the relationship between postoperative VF defects RNFL damage. Design: Retrospective interventional case series. Participants: Thirty-four eyes of 32 patients with idiopathic macular holes that underwent vitrectomy January 2001 and March 2003 were included in this study. Eyes were divided into 3 groups according to the occurrence of postoperative VF defects and the use of indocyanine green for internal limiting membrane peeling during surgery: 11 eyes with VF defects after indocyanine green-assisted vitrectomy (group 1), 9 eyes without VF defects despite the use of indocyanine green (group 2), and 14 eyes without VF defects that underwent vitrectomy without indocyanine green (group 3). Methods: Retinal nerve fiber layer thickness in e Results: The mean RNFL thickness in 3 of 4 quadrants (superior, nasal, inferior) was measured with OCT. Main Outcome Measure: Retinal nerve fiber layer thickness around the optic disc. group 1 was significantly less than that in the corresponding quadrant in groups 2 and 3 (P <0.01). In the temporal quadrant, there was a significant difference between groups 1 and 3 (P = 0.02), but not between groups 1 and 2 . The RNFL thickness in group 1 was significantly less in 3 quadrants in operated eyes than in fellow eyes (P <0.05). Conclusions: The RNFL thickness was reduced in eyes with VF defects after indocyanine green-assisted vitrectomy for macular holes, suggesting that the postoperativeVF defects have been caused by RNFL damage relating to the use of indocyanine green.