To evaluate the performance of calcium suppressed images (CaSupp) obtained by dual-layer detector computed tomography (DLCT) for the detection of bone marrow edema (BME) in patients with wrist pain.

We retrospectively analyzed 49 patients with wrist pain (44 distal radius fractures, 2 carpal bone fractures, 2 scaphoid nonunion advance collapses, 1 Kienböck disease), who underwent both DLCT and MRI. Two blinded and independent readers evaluated CaSupp images for evaluating BME by using color-coded maps. Using MRI images as the reference standard, the sensitivity and specificity of CaSupp images were analyzed for detecting BME of radius, ulna, and carpal bones.

On MRI, 44 distal radius and 30 distal ulna fractures were found. In detecting BME of radius and ulna, two readers showed 100% of agreement. When CaSupp images were compared with MRI images, sensitivity and specificity for detecting BME were both 100% for radius, and 88% and 87.5% for ulna, respectively. For carpal bone, BME was found in 8 of 44 radius fractures and 5 of patients with only carpal bone abnormalities on MRI. Those carpal bone BMEs were detected on CaSupp images with following diagnostic accuracy: sensitivity, 92.8% for reader 1 and 64.2% for reader 2; specificity, 88.5% in both readers. For detection of carpal bone BME, two readers showed moderate agreement (agreement 75.5%, kappa value 0.43).

CaSupp images reconstructed from DLCT enabled detection of BME in fractured distal radius and ulna with substantially high diagnostic accuracy when compared to MRI images. However, CaSupp demonstrated limited performance in visualization of BME of carpal bone pathologies.

CaSupp images showed similar performance in visualization and detection of BME in wrist, including incomplete fracture compared with MRI. CaSupp images is expected to be a promising technique to demonstrate BME in wrist.