Imaging signs for preoperative diagnosis of dural ossification in patients with thoracic ossification of the ligamentum flavum: a blind, randomized diagnostic accuracy study.

Abstract

Dural ossification (DO) is the leading cause of surgery-related dural tear in patients with ossification of the ligamentum flavum (OLF). An accurate preoperative diagnosis of DO is conducive to the selection of appropriate surgical methods. Although several imaging signs, such as Banner cloud sign (BCs), tram-track sign (TTs), and comma sign (Cs) have been proposed for the preoperative diagnosis of DO, their diagnostic value has not been well studied. The aim of this study was to explore the diagnostic value of BCs, TTs, and Cs, and provide evidence-based data for their clinical application. This is a blind, randomized diagnostic study using retrospectively collected data from 102 consecutive patients who were diagnosed with OLF and underwent decompression surgery between January 2018 and June 2019. A total of 8 surgeons with different qualifications were recruited to read these imaging signs to identify the presence of DO. Surgical records were used as the reference standard. Sensitivity, specificity, and the area under the receiver operating characteristic (ROC) curve (AUC) were used to evaluate the diagnostic accuracy of each imaging sign and their different combinations. Of the 102 patients, 21 were diagnosed with DO. BCs had a significantly higher diagnostic accuracy than TTs and Cs, with the AUC of 0.704, 0.607, and 0.593, respectively. The specificity of BCs, Cs, TTs, and their combination in diagnosing DO was 91.5%, 92.1%, 68.3%, and 62.2%, respectively. In the combined diagnostic test, the results showed that the combined diagnosis accuracy of BCs and Cs was the highest, and the AUC was 0.738. The combination of BCs, Cs, and TTs increased the sensitivity of diagnosing DO (77.5%), but did not improve the diagnostic accuracy, and the AUC was 0.699. BCs had higher diagnostic accuracy than TTs and Cs. BCs and Cs were highly specific for DO, whereas TTs could be confusing due to their non-specific presentations. The combination of BCs, TTs, and Cs improved the sensitivity of DO diagnosis, but not the specificity and accuracy.