The diagnosis of Marfan syndrome (MFS) is defined by a combination of major and minor criteria, related to the different systems involved, according to the Ghent nosology of the spine. Spinal imaging can detect both skeletal (including scoliosis and spondylolisthesis) and neurological involvement (i.e. dural ectasia). The aim of the present study was to assess the interest of screening the rachis by conventional radiography CR and complementary imaging (computed tomography [CT] or magnetic resonance imaging [MRI]) in patients suspected of MFS, and to modelise the most relevant imaging procedure to diagnose MFS.

Evaluation of the sensitivity and specificity of CR of the lumbosacral spine versus sectional imaging for the detection of dural ectasia (DE) in a subgroup of 92 patients suspected of MFS. Retrospective analysis of the contribution of CR to the diagnosis of MFS in 1992 patients referred to our clinic.

DE was detected by CR in 12 of the 92 patients (13%) and was always confirmed by CT or MRI. Complementary imaging alone detected 33 DE (35.9%). All patients with DE detected by CR were diagnosed with MFS. Among the 1992 patients, 591 were confirmed MFS; 117 patients had DE detected by CR (19,8%) while 12 (2,0%) were detected by complementary imaging. In MFS patients, 98 (16.6%) had significant scoliosis and 14 (2.4%) had spondylolisthesis. The positive predictive value of DE detected by CR for the diagnosis of MFS was 92.9% (95% IC: 86.8–96.4), and the negative predictive value was 74.6% (95% IC: 72.6–76.5). We conclude that spinal imaging is useful for the diagnosis of MFS.