Table 2.
Virtual non-calcium computed tomography (CT) ability to depict traumatic bone marrow edema on dual-energy spine scans
| Authors, year [reference] | DECT platform | Number of patients | Qualitative analysis (sensitivity, specificity, accuracy) | Quantitative analysis cutoff (sensitivity, specificity, accuracy) | Reference standard | Spine site |
|---|---|---|---|---|---|---|
| Wang et al., 2020 [51] | DSCT | 20 | 85%, 97%, 93% | -12 HU (95%, 86%, 98%) | MRI | T/L |
| Jeong et al., 2020 [52] | DSCT | 31 | 83%, 99%, 99% | NA | MRI | T/L |
| Booz et al., 2020 [29] | DSCT | 52 | 93%, 95%, 90% | -43 HU (85%, 95%, 97%) | MRI | S |
| Foti et al., 2019 [58] | DSCT | 76 | 88%, 92%, 90% | cutoff NA (92%, 90%, 91%) | MRI | T/L |
| Diekhoff et al., 2019 [53] | SACT | 70 | 72%, 70%, NA | NA | MRI | T/L |
| Frellessen et al., 2018 [45] | DSCT | 51 | 96%, 96%, 98% | NA | MRI | T/L |
| Diekhoff et al., 2017 [57] | SACT | 9 | 88%, 100%, NA | NA | MRI | T/L |
| Petristsch et al., 2017 [54] | DSCT | 22 | 64%, 99%, 93% | -47 HU (92%, 82%, 84%) | MRI | L |
| Kaup et al., 2016 [55] | DSCT | 49 | 90%, 90%, 95% | NA | MRI | T/L |
| Bierry et al., 2014 [10] | DSCT | 20 | 84%, 97%, 95% | cutoff NA (85%, 82%, NA) | MRI | T/L |
| Wang et al., 201 3[27] | DSCT | 63 | NA | -80 HU (96%, 98%, 97%) | MRI | T/L |
DECT Dual-energy computed tomography, DSCT Dual-source computed tomography, SACT Sequential acquisition computed tomography, HU Hounsfield units, NA Not available, MRI Magnetic resonance imaging, T Thoracic, L Lumbar, S Sacral