Usefulness of the indocyanine green fluorescence endoscope in endonasal transsphenoidal surgery

Abstract

OBJECT To avoid disorientation during endoscopic endonasal transsphenoidal surgery (ETSS), the confirmation of anatomical landmarks is essential. Neuronavigation systems can be pointed at exact sites, but their spatial resolution power is too low for the detection of vessels that cannot be seen on MR images. On Doppler ultrasonography the shape of concealed arteries and veins cannot be visualized. To address these problems, the authors evaluated the clinical usefulness of the indocyanine green (ICG) endoscope. METHODS The authors included 38 patients with pituitary adenomas (n = 26), tuberculum sellae meningiomas (n = 4), craniopharyngiomas (n = 3), chordomas (n = 2), Rathke's cleft cyst (n = 1), dermoid cyst (n = 1), or fibrous dysplasia (n = 1). After opening the sphenoid sinus and placing the ICG endoscope, the authors injected 12.5 mg of ICG into a peripheral vein as a bolus and observed the internal carotid arteries (ICAs), cavernous sinus, intercavernous sinus, and pituitary. RESULTS The ICA was clearly identified by a strong fluorescence signal through the dura mater and the covering thin bone. The intercavernous and cavernous sinuses were visualized a few seconds later. In patients with tuberculum sellae meningiomas, the abnormal tumor arteries in the dura were seen and the vague outline of the attachment was identified. At the final inspection after tumor removal, perforators to the brain, optic nerves, chiasm, and pituitary stalk were visualized. ICG fluorescence signals from the hypophyseal arteries were strong enough to see and spread to the area of perfusion with the passage of time. CONCLUSIONS The ICA and the patent cavernous sinus were detected with the ICG endoscope in real time and at high resolution. The ICG endoscope is very useful during ETSS. The authors suggest that the real-time observation of the blood supply to the optic nerves and pituitary helps to predict the preservation of their function.