Abstract
Purpose: The nasal bone is the most frequently fractured facial bone. Discordance between the CT reading by the radiologist and the diagnosis by the plastic surgeon is not uncommon. This study examined the discordance and proposes a method for reducing the rate of discordance. Methods: The CT readings by the radiologist were compared with the diagnosis by the plastic surgeon in 716 patients with a clinically suspected nasal bone fracture. The CT reading was classified as the following: a nasal bone fracture, suspicious nasal bone fracture, old nasal bone fracture, no nasal bone fracture. The sensitivity, specificity and positive predictive value of the CT reading were calculated. Results: A nasal bone fracture was diagnosed in 646 patients by the plastic surgeon and confirmed intraoperatively. The reading of a "nasal bone fracture", "suspicious nasal bone fracture", "old nasal bone fracture" and "no nasal bone fracture" was 85.8%, 4.6%, 0.6% and 9.1% respectively. The sensitivity and specificity of the CT reading were 95.0% and 92.9%, respectively. The positive predictive value of the CT reading was 99.3%. The reading of "nasal bone fracture" that was not a nasal bone fracture clinically was 17.1% (12 of 70), and the reading of "no nasal bone fracture" or "old nasal bone fracture" that was found to be a nasal bone fracture clinically was 3.3% (21 of 646). The discordance rate between the CT reading by the radiologist and the diagnosis by the plastic surgeon was 4.6%. Conclusions: To reduce the discordance rate, we propose to hold a meeting with the plastic surgery-radiology staff to communicate the information regarding a suspicious or old nasal bone fracture.