LAUSR

BACKGROUND In some fluoroscopy machines, the dose-rate output of the fluoroscope is tied to a selectable patient size. Although patient size may play a significant role in visceral or cardiac procedures, head morphology is less variable, and high dose outputs may not be necessary even in very obese patients. We hypothesized that very small patient size setting can be used to reduce dose for cerebral angiography without compromising image quality. METHODS Patients who underwent endovascular neurosurgical procedures during the 2015-2016 academic year were identified, and estimated procedural air kerma (AK) was tabulated retrospectively. Technologists were instructed to begin using the very small patient size setting for all procedures performed using our Philips Allura Xper FD20 biplane fluoroscopy system beginning in March 2016. No changes were made in a second procedure room using a Toshiba Infinix system. Student t tests and logistic regression models were used to compare radiation exposure before and after March 1, 2016, for both machines. RESULTS For diagnostic cerebral angiograms performed on the Philips system (n = 302), AK was reduced by approximately 17% (1277 vs. 1061 mGy; P = 0.0006.) Changes in table height, total fluoroscopy time, patient weight, and body mass index did not contribute to this difference. No significant change was seen in total AK using the Toshiba system (n = 237). Blinded review by a neuroradiologist did not demonstrate any change in image quality. CONCLUSIONS Using the very small patient size reduces fluoroscopy dose by 17% for cerebral angiography without impacting image quality.