To the Editor:—
I read with interest the results of the American Society of Anesthesiologists Postoperative Visual Loss Registry and the analysis of the 93 spine surgery cases with postoperative visual loss.1Striking to me was that the majority of the complications happened in settings that were thought safe in the past. It has long been taught that prevention of direct ocular pressure, severe hypoxia, anemia, and hypotension prevent blindness in the majority of patients undergoing prone spine surgery. This report and analysis of data showed that direct ocular pressure contributed to only a small percentage of the documented cases, and that blindness occurred over a wide range of systolic pressure, homodynamics, and hemoglobin concentrations. That led me to conclude that while prevention is the best cure for this problem, best prevention is not currently understood; it raised in my mind the question of intraoperative visual system monitoring. Today, we routinely use pulse oximetry, capnography, and even processed electroencephalographic monitoring to identify and promptly correct hypoxemia, ventilatory inadequacy, and awareness. Isn’t it logical that in high-risk cases where blindness is possible that we should be monitoring the patient intraoperatively to identify early retinal changes that could correlate with this tragic event and try to prevent that outcome?
Intraoperative retinal monitoring through visual evoked potentials, with the aim of preserving visual fields, has been used successfully in many cases such as intracranial surgeries,2occipital corticectomy for epilepsy,3functional endoscopic sinus surgery,4optic nerve function surgery, and other surgeries involving the visual pathway. Currently, the use of the visual evoked potential is limited and not routinely practiced in spine surgery performed in prone positioning. It seems obvious that this modality should be used more frequently and even routinely in all prone spine surgeries.
I am aware of conflicting reports about the usefulness of this monitoring modality, but I believe that our reading and correlation of retinal evoked potentials will improve as the monitoring becomes routine. I hope that the future will focus on improving monitoring of the visual evoked potential, perhaps in a form as simple as bispectral monitoring (such as the Bispectral Index®; Aspect Medical Systems Inc., Norwood, MA). Such monitoring may allow us to accurately detect early, reversible damage to the visual pathway and enable us to prevent permanent problems. This would be in the best tradition of anesthesiology.
Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio. firstname.lastname@example.org