Optical coherence tomography for optic disc edema.

and the temporal association and clinical presentation indicate the air infusion as a likely cause. In the second and third case, clinical parameters were altered in a similar fashion during air/fluid exchange, and the cardiac auscultation revealed millwheel murmur in both cases, which is highly indicative of VAE. In both cases, no other causes could be elucidated from the anesthetic point of view. The temporal association and clinical presentation again suggest air infusion as a likely cause. In the first and third case, when the vitreous cavity was returned to a fluid-filled state, the clinical parameters improved, and the patient recovered. All of this suggests that air/fluid exchange during vitrectomy has the potential to cause venous air embolism. The criterion standard of confirming air embolism is through transesophageal echocardiography. However, this equipment is not readily available intraoperatively, especially in the setting of vitreo-retinal surgery. We agree with El-Annan and Barr that one would normally not expect air to enter a vortex vein or orbital vein, but it is conceivable that, in the setting of trauma, those veins might be torn open in a way that allows the air to enter. The third case suggests that this might even be possible in a nontraumatized eye if the cannula is in the choroidal space and the air dissects posteriorly and tears open a vortex vein. We came across these cases during a review of the literature for a different subject, and we recognized that these isolated anesthesia cases had never been described in the ophthalmic literature. Our concern was that clinicians would not know to consider this complication if a patient began to decompensate shortly after air fluid exchange. Because the most important intervention is to discontinue the air infusion, and because the responsibility for doing this rests with the ophthalmologist, we felt it important to make others aware that the possibility for this potentially fatal problem exists.