Abstract
BACKGROUND: Atlantodental subluxation (ADS) is a serious condition that can result in sudden death. Measuring the anterior atlantodental interval (AADI method) is the gold standard for diagnosis but the complex anatomy of this region can make diagnosis difficult, especially for beginners. Therefore, we would like to use a simpler method, the Swischuk line method, to diagnose ADS. The purpose of our study was to evaluate the diagnostic performance of the Swischuk line method for ADS on lateral cervical spine radiographs compared to the AADI method. METHODS: A retrospective study was conducted with patients who presented with ADS (ADS group, n = 32, mean age 57.78 years, age range 34-82 years, 10 men, 21 women) and an age- and sex-matched control group (n = 32). The diagnostic performance of the AADI method and the Swischuk line method for ADS was assessed using lateral cervical radiographs in both flexion and neutral postures by an experienced musculoskeletal radiologist (reader 1), a senior resident (reader 2), and a junior resident (reader 3) in the radiologic department. RESULTS: In the flexion posture, the AADI method and the Swischuk line method showed excellent diagnostic performance with AUCs > 0.9 for readers 1 2 and reader 3. In a neutral posture, the diagnostic performance of the AADI and Swischuk line methods was decreased. With a 1 mm cut-off value using the Swischuk line method in flexion posture, the sensitivity was 75% or more, the specificity was 100%, and the accuracy was 87.50% or more 90.63% for all readers. With a 2 mm cut-off value, the sensitivity was low (37.50-46.88%) but the specificity was 100% for all three readers. In a neutral posture, the sensitivity for both methods decreased, though specificity remained high (> 80%). CONCLUSIONS: The Swischuk line method was found to be reliable and showed high sensitivity and specificity with a cut-off value of 1 mm for the diagnosis of ADS in cervical lateral radiographs in flexion posture. It can be used as a complement to the AADI method.