Dural ossification associated with cervical ossification of the posterior longitudinal ligament: frequency of dural ossification and comparison of neuroimaging modalities in ability to identify the disease
Object. The authors' goal in this study was to understand the frequency and pattern of dural ossification (DO, and to evaluate the effectiveness of neuroimaging modalities used to identify this disease in association with ossification of the posterior longitudinal ligament (OPLL). Methods. One hundred eleven patients with OPLL underwent anterior procedures. Of these patients, 17 (15.3%) had associated ossification of the dura mater. There were 10 cases of DO in the 94 patients with segmental OPLL and seven in the 17 patients with nonsegmental OPLL (seven continuous and 10 mixed-type OPLL). Retrospective evaluation of DO was performed by examining plain x-ray films, polytomography studies, computerized tomography (CT) scans, and magnetic resonance (MR) images. A positive correlation was found between the type of OPLL and the frequency of DO (p < 0.01). The DO was classified into the following three types according to shape: 1) isolated type, 2) double-layer type, and 3) en bloc type, based on its relationship with OPLL. There were 10 lesions of the double-layer type, four en bloc type, and three isolated type; the double-layer pattern of DO was the most common. All DOs as well as OPLLs (17 cases) were identified using bone-window CT scanning. Polytomography was used successfully to identify all 12 OPLLs, whereas DO was recognized in seven of the 12 cases. Magnetic resonance imaging could not identify DO (none of the 17 cases), although OPLL was identified on MR imaging in 12 of the 17 cases. Conclusions. Three patterns of DO associated with OPLL, that is, isolated, double-layer, and en bloc types, were confirmed by retrospective analysis of neuroimaging findings. Nonsegmental OPLL was likely to be accompanied by DO. Bone-window CT scanning was most useful for identification of DO as well as OPLL, whereas MR imaging was ineffective in recognizing DO.