Assessing the accuracy of ICG videoangiography in predicting visual outcomes for anterior skull base surgeries, compared to intraoperative visual evoked potential

Introduction: Anterior skull base tumors possess the potential to infiltrate or exert pressure on the optic nerve, which may lead to visual impairment resulting from ischemia or demyelination caused by impaired blood supply to the optic nerve (ON). The preservation of microcirculation within the ON during decompression procedures is of utmost importance. However, a dependable intraoperative assessment tool for forecasting visual outcomes remains to be established. This study aims to present utilizing indocyanine green videoangiography (ICG-VA) as a predictive modality, in comparison with intraoperative visual evoked potentials (VEP), for the first time. Methods: Thirteen patients with perichiasmatic pathologies were selected for analysis. Each underwent advanced imaging, including flow analysis with Flow 800 software (Carl Zeiss). The intervals from fluorescence appearance in the internal carotid artery (ICA) to its presence in the pial circulation of the ON, and the full saturation of both, were carefully measured before and after resection. The control cohort included 5 elective aneurysm clipping procedures in the anterior part of the Circle of Willis. Results: A total of 13 patients and five control subjects (mean age: cases 55.30±15.2 years, controls 57.8±2.38, p<0.05) were included. Twelve patients experienced impaired visual fields preoperatively, 10 showed improvement following surgery. Prolongation of ipsilateral P100 latency was observed in 11 tumor patients and 3 control patients on the ipsilateral side, 9.81±13.11 msec and 5.9±8.33 msec for the tumors and control patients, respectively. The sensitivity of VEP was 60%, with a specificity of 50%, a positive predictive value of 85.7%, and a negative predictive value of 20%. In contrast, the sensitivity for ICG-VA was 44%, with 100% specificity, a 100% positive predictive value, and a 28.5% negative predictive value. The mean improvement in the peak time of the optic nerve relative to the ipsilateral internal carotid artery was 0.82±2.32 seconds after lesion resection. The peak time difference ICA-ON for control patients was 0.45±0.87 sec (p<0.05). The accuracy rates were 63% for ICG-VA and 58% for VEP. Conclusion: ICG-VA of the optic nerve presents a promising modality for predicting visual outcomes in comparison to VEP. However, additional research and data are required to establish a significant difference.