Intraoperative Assessment of Epiretinal Membrane Peeling

All interventions were performed by two experienced vitreoretinal surgeons (T.C., A.B.) following local anesthesia (peribulbar block) and antisepsis with povidone–iodine solution. Both groups received a standard 25-Gauge three-port vitrectomy (Constellation^® Vision System, Alcon Laboratories Inc, Fort Worth, TX, USA) followed by a staining of the epiretinal tissues by the injection of a combination 0.15% TrypanBlue, 0.025% Brilliant Blue G and 4% polyethylene glycol vital dye (Membraneblue-Dual^®, DORC International, Zuidland, The Netherlands). Prior to epiretinal membrane peeling, a small bubble of approximately 1.0 cc of Perfluoro-N-Octane (EFTIAR Octane^®, DORC International, Zuidland, The Netherlands) was injected over the macula in the PFCL-assisted group.
During the intervention, upon creating a stable flap of epiretinal tissue with end-gripping forceps, the surgeon continued applying a mild traction and proceeded to acquire an iOCT B-Scan (Rescan 700^®, Carl Zeiss Meditec AG, Jena, Germany) oriented in parallel to the projection of the vector of traction over the retina (Figure 1a). The acquired scans had a width of 6 mm and an A-Scan depth of 5.8 mm in tissue. The same imaging procedure was repeated whenever the flap was dragged into a different macular subfield or a new flap was started (Figure 2). Superior, nasal, inferior and temporal macular subfields were estimated following a simplified version of the ETDRS grid (Figure 1b). The careful evaluation of intraoperative complications such as the occurrence of retinal tears was conducted. Moreover, an independent observer (G.G.) collected intraoperative data regarding the number of grabs the surgeon had to perform with the forceps in order to complete the ERM removal.
Following a complete peeling of the ERM, the surgeon further proceeded to completely remove any residual ILM, with additional injections of vital dye as needed. In the PFCL-assisted group, the PFCL was then thoroughly removed by active aspiration and fluid/air exchange was performed in both groups.
All OCT images for each eye were exported to external software (ImageJ, NIH, USA; version 1.53e) for postoperative analysis (Figure 1c). Two masked independent ophthalmologists (S.M.P., G.G.) assessed the DA value for individual OCT scans (Figure 3) and then reported the mean value expressed in degrees for each eye for statistical analysis. A minimum of four scans—at least one for each macular subfield—per eye were examined. The interclass correlation coefficient (ICC) was determined to assess interobserver agreement. The statistical analysis of the results was conducted via the IBM SPSS Statistics^® software (version 25).