Intraoperative monitoring of visual evoked potentials: experience of 240 operations

Abstract

Background. Intraoperative monitoring (IOM) of visual evoked potentials (VEPs) is used to inform surgeons about impacts on the visual system in order to prevent iatrogenic visual impairment. The VEP monitoring use become widespread only in the last decade; nowadays, there is no generally accepted methodology for its implementation, and the effectiveness of VEP monitoring and the factors determining it have not been sufficiently studied.Aim. The aim of the study was to investigate the factors influencing the VEP monitoring feasibility and effectiveness.Materials and methods. Data from 240 consecutive neurosurgical operations performed using VEP monitoring were retrospectively reviewed. IOM data (registration parameters, presence and type of VEP changes), patient characteristics (gender and age, tumor type and location, presence of preoperative visual dysfunctions), anesthesia parameters and postoperative changes in vision were studied. Statistical analysis was performed using χ2 and Mann–Whitney tests.Results. VEPs were obtained in 91.3 % of eyes with completely or partially preserved vision. The main factors reducing the chances to record VEPs successfully are preoperative visual disorders and the use of inhalation anesthesia. A personalized approach to the selection of reference electrodes and frequency filtering parameters makes it possible to reduce the number of averagings required for VEP recording and accelerate informing surgeons. With successful monitoring 59.1 % of eyes had no noticeable VEP changes; 5.8 % of eyes had signs of intraoperative improvement; 35.1 % had signs of deterioration. Among the last category, 60.7 % of eyes had full VEPs recovery afterwards. After surgery, new visual disorders were detected in 2.6 % of eyes without signs of intraoperative deterioration, in 6.7 % – with temporary deterioration, and in 19.3 % – with signs of deterioration persisted until IOM is finished. Assessing the sensitivity and specificity of VEP monitoring is hampered by the possibility of complications in the early postoperative period and IOM influence on the course and results of the operation. The proportion of total resections was maximal when VEP monitoring was successful. In the subgroup without preoperative visual impairments, the alarms during monitoring were associated with decrease in proportion of total resections proportion due to increase in proportion of subtotal resections.Conclusion. VEP monitoring with a personalized approach allows effective monitoring of visual functions preservation during neurosurgical operations.