To the Editor:—
The American Society of Anesthesiologists (ASA) Closed Claims Project has provided valuable information regarding risks and potential etiologies of untoward events related to the practice of anesthesia.1The ASA Registry for Postoperative Visual Loss arose from some of the same concerns as did the Closed Claims Project: an attempt to understand problems that have become medical–legal issues and to provide better care for our patients.2Lee et al. 3have provided a valuable service in documenting data associated with this rare and devastating adverse event. Their report follows closely the recent ASA “Practice Advisory for Perioperative Visual Loss Associated with Spine Surgery.”4Inasmuch as randomized prospective clinical trials to discern etiology and efficacy of the suggested therapies of increasing blood pressure and hemoglobin concentration would not be feasible, owing to the low incidence, estimated to be approximately 0.03–0.1% for ischemic optic neuropathy (ION)5,6(a reduction of 25% would require a study of approximately 200,000–750,000 patients per group), other methodologies are needed to assess possible etiologies and therapies. As pointed out by Lee et al. , unfortunately, information regarding the total number of surgical procedures represented by the reports in their database is not available. The registry could be improved by asking those who provide case reports to also indicate the number of similar operations performed during a several-year period (a short period would produce an artificially estimated high incidence). Even this, however, would overstate the incidence, because this complication has never been encountered by most spine surgeons,7and likely most institutions. Of greater concern is the recommendation contained in the report and the absence of other recommendations.
We question the recommendation regarding routine preoperative discussion of the possibility of postoperative visual loss, given the exceedingly low incidence. Complications of such low incidence5,6(e.g. , masseter muscle rigidity/malignant hyperthermia8,9) are not routinely discussed, and the rarity of ION makes it unlikely that discussion would be a relevant consideration in whether the patient elected to proceed. In addition, once mentioned, little can be said regarding prevention or therapy, inasmuch as the etiologies of anterior ION and posterior ION are uncertain, and prophylactic and therapeutic maneuvers are of unproven value.
Of interest are the surprising data that the patients’ eyes were documented as having been checked in only 51% of cases of ION (frequency not given) and in just 6 of 10 cases of central retinal artery occlusion (frequency of between every 30 min and only once during the entire procedure), which is widely regarded as being caused by direct trauma or pressure applied to the eye. Our spine anesthesia team was established in 1991, and our routine care includes checking the eyes every 15 min of every patient in the prone position. We previously reported 7 cases of visual disturbances after 3,450 spinal surgeries, including four IONs, one central retinal venous thrombosis, and no central retinal arterial thromboses.5We are surprised that the registry report contained no recommendation regarding the advisability of frequent checks for absence of direct pressure on the patient’s eyes: something that is easily performed, is of no cost, and makes sense physiologically, although of unproven efficacy in preventing central retinal artery thrombosis. In addition, we recommend a simple, quick test of crude visual function and visual fields (e.g. , tell how many fingers, and when they can be seen as they are moved from the periphery to a central position) as soon as possible in the immediate postoperative period. The ASA practice advisory4and Myers et al. 7in their evaluation of a series of 37 cases of visual loss after spinal surgery also recommend an early postoperative assessment of visual function. This allows for rapid consultation, documentation of the timing of the event, and institution of any recommended, although unproven, therapy.
The report provides a good discussion regarding possible etiologies of ION, including increased venous pressure and trapping of the optic nerve owing to increased interstitial fluid accumulation and thus pressure in an enclosed bony canal. It is possible that the latter issue may also decrease arterial blood flow. As discussed in the report, placing a patient prone in a position with the head slightly elevated decreases intraocular venous pressure. We practice and recommend this, as does the ASA practice advisory for “high-risk patients.”4In addition, we also limit the volume of crystalloid solution to reduce the possibility of increased interstitial fluid and pressure, although, admittedly, neither this nor the slightly head-up tilted position is a proven efficacious prophylactic therapy.
We were surprised that the report did not consider patients’ fraction of inspired oxygen or arterial oxygen tension. We have shown that anemia-induced neurologic deficits in healthy people can be reversed by increasing arterial oxygen concentration.10We are further concerned that both Lee et al. and the ASA Task Force suggest that protracted surgery and amount of blood loss are risk factors for the development of postoperative visual loss. Neither is physiologically grounded. A more sensible assessment, in the absence of a validated monitor for visual function during anesthesia, would focus on blood loss replacement and maintenance of normovolemia, rather than the volume of loss itself, and the duration of factors that might influence inadequate perfusion of the ophthalmic vasculature, rather than the duration of the surgery. The latter may be a poorly correlating surrogate for hypovolemia/hypoperfusion and may appear erroneously as a univariate factor in a database of a limited number of events. These might also be surrogates for the intravenous infusion of substantial amounts of salt solutions, with the potential adverse action noted above. Anesthesiologists and surgeons should work together to minimize potential contributing factors to this devastating complication; however, in the absence of definitive data, the Task Force’s suggestion to alter accepted surgical practice11,12is questionable.
*University of California, San Francisco, California. rwes@novonordisk.com
