In Reply:—
In response to Dr. Larson’s rather dogmatic conclusions on how to avoid perioperative ischemic optic neuropathy, I am pleased that he has never personally experienced this complication in one of his patients. His observation drives home the primary point of the report by Dr. Lee et al. 1and my editorial2: There are too few of these complications at this time to scientifically deduce causative risk factors. Quite simply, it is not logistically or financially possible at this time to prospectively search for causative risk factors of this devastating complication as it occurs in patients undergoing spine surgery while positioned prone.
Therefore, it is difficult to understand what data Dr. Larson uses as a basis for his recommendations. There are no data to suggest that limiting crystalloid administration to less than 40 ml/kg regardless of duration of the surgical procedure impacts ischemic optic neuropathy (negatively or positively). The same can be said for his suppositions about hematocrit levels of less than 26 and limiting durations of controlled hypotension to only the dissection period of spine surgery. Data from multiple studies document that many patients who have Dr. Larson’s “risk factors” do not develop ischemic optic neuropathy—and many who develop ischemic optic neuropathy receive crystalloid volumes of less than 40 ml/kg, have hematocrits intraoperatively well above 26, and are provided care without the use of controlled hypotension. In short, there is no scientific reasoning to justify Dr. Larson’s strongly worded, unsupportable recommendations.
Dr. Weiskopf raises two points to which I would like to respond. First, he speculates that periodic intraoperative checks of the eyes for absence of direct pressure on patients’ eyes may be useful in preventing central retinal artery thrombosis. His spine team evidently established periodic intraoperative eye checks for all prone-positioned spine surgery patients and found that none of their 3,450 patients developed this complication.3However, as he notes, the frequency of this event is very low. It is, therefore, impossible to draw any conclusion or even inference that his team’s eye checks had anything to do with the outcomes that their patients experienced. Regarding the use of eye checks, it is disappointing to find that 6 of the 10 patients with central retinal artery occlusion in the America Society of Anesthesiologists Visual Loss Registry had at least one eye check during their procedures.1In those 6 patients, eye checks apparently did not prevent this problem from occurring. Therefore, should eye checks be done? Intuitively, yes—they are certainly cheap and easy. However, there are no data showing that they are effective in reducing the frequency or severity of central retinal artery occlusion.
Dr. Weiskopf also indirectly suggests that increased inspired oxygen concentrations, resulting in elevated arterial oxygen tension, may be helpful in decreasing the frequency of perioperative vision loss. He cites an excellent study that he and colleagues performed in volunteers who were made acutely anemic, finding that increased inspired oxygen concentrations reversed the negative cognitive effects of the anemia.4Ironically, hyperoxia has an autoregulation-related vasoconstrictive effect on the end-retinal arterioles of the central retinal artery, reducing both the diameter of the arterioles and also their blood flow and velocity.5It is not clear what impact this finding has on blood flow to the various regions of optic nerves, but increasing inspired oxygen concentrations may not be as helpful as he seems to suggest.
Mayo Clinic, Rochester, Minnesota. warner.mark@mayo.edu
