Purpose: To evaluate the possible toxicity of indocyanine green (ICG) dye in macular hole surgery by comparing functional results after successful surgeries performed with and without staining.
Design: Prospective observational case series.
Methods: In a study conducted at our department, 30 eyes of 29 patients with idiopathic macular hole underwent pars plana vitrectomy, with peeling of the internal limiting membrane (ILM). In 21 eyes (group A), ICG staining was used to visualize the membrane; in nine eyes (group B), no dye was applied. Pre- and postoperative examinations (at three, six, 12, and 20 months) included visual acuity (VA) (Snellen chart), slit-lamp biomicroscopy, multifocal electroretinography (mfERG), and optical coherence tomography (OCT). For statistical analysis, the Mann-Whitney U test and Newman-Keuls post hoc analysis were used.
Results: The macular hole was closed in all patients. At 20 months, VA improvement (logarithm of minimal angle of resolution units) compared with baseline was more pronounced in group B (P < .001) than in group A (P = .022); VA was also better for group B compared directly with group A (P = .048). For mfERG, preoperative responses were subnormal. Postoperatively, responses initially decreased, but at 20 months significant improvement was seen in both groups (P < .001). When data from the groups are compared at 20 months, significantly greater mfERG improvement was found in group B in both central and perimacular areas (P < .001).
Conclusions: The significantly better outcomes without staining may suggest dye toxicity. Alternative internal limiting membrane visualization substances with fewer adverse effects (trypan blue, triamcinolone) may be preferable.