Fixation stability and implication for multifocal electroretinography in patients with neovascular age-related macular degeneration after anti-VEGF treatment

Purpose: To quantify fixation stability in patients with neovascular age-related macular degeneration (nAMD) at baseline, 3 and 6 months after anti-vascular endothelial growth factor (anti-VEGF) treatment and furthermore asses the implications of an unsteady fixation for multifocal electroretinography (mfERG) measurements. Methods: Fifty eyes of 50 nAMD patients receiving intravitreal anti-VEGF treatment with either bevacizumab or ranibizumab and eight eyes of eight control subjects were included. Fixation stability measurements were performed with the Eye-Link eyetracking system and the retinal area in degrees² (deg²) containing the 68 % most frequently used fixation points (RAF68) was calculated. MfERG P1 amplitude and implicit time were analyzed in six concentric rings and as a summed response. Patients were examined at baseline, 3 and 6 months. Four different mfERG recordings were performed for the control subjects to mimic an involuntary unstable fixation: normal central fixation, 2.4°, 4.8°, and 7.1° fixation instability. Results: For control subjects, a fixation instability of 2.4° (corresponding to the central hexagon) did not reduce mfERG ring amplitudes significantly, whereas 4.8° and 7.1° fixation instability reduced the amplitudes significantly in rings 1 and 2 (p <0.001) as well as in the peripheral rings in the 7.1° instability condition (p <0.001). Fixation stability improved non-significantly for patients at 3 and 6 months. The size of the retinal area of fixation was at baseline, 3 and 6 months negatively correlated to visual acuity (VA) (r_baseline = −0.65, r_{3 months} = −0.60, and r_{6 months} = −0.66 respectively, p <0.001) and mfERG amplitudes of the three innermost rings (r_baseline = −0.29, p = 0.042, r_{3 months} = −0.43, p = 0.003 and r_{6 months} = −0.31, p = 0.042). The VA cutoff for a fixation area less than 5 deg² (approximately the central hexagon) was 65, 77, and 68 ETDRS letters (corresponding a maximal Snellen equivalent of 0.31) at baseline, 3 and 6 months, respectively. Conclusions: MfERG amplitudes in recordings of nAMD patients are at substantial risk of being reduced due to poor fixation as a large number of patients may use a fixation area of more than 5 deg². Fixation monitoring during recording as well as interpretation of results should be performed with care, especially in patients with poor visual acuity.