Comparison | MR scanner (T) | Patients | Patients, n | Joint/bone | BME feature | Correlation coefficient(P values) | Accuracy/sensitivity | |
Schmid et al 9 | STIR versus T1Gd | 1.0 | Various orthopaedic diagnoses (osteoarthritis, osteonecrosis, insufficiency fractures and the majority (n=32) were non-specific cases | 51 | Foot and ankle | BME volume | 0.98(p<0.001) | nd |
Mayerhoever et al 8 | STIR versus T1Gd | 1.0 or 1.5 | Various orthopaedic diagnoses (bone bruises, osteoarthritis, osteonecrosis, stress fractures) | 30 | Knee | BME volumeSignal contrast | 0.99(p<0.001)0.94(p<0.001) | nd nd |
Tamai et al 10 | STIR versus T1Gd | 1.5 | Early RA | 51 | Wrists,MCPs,PIPs | Presence of BME | nd | Accuracy 98% |
Stomp et al 7 | T2 fatsat, versus T1Gd | 1.5 | Early arthritisAdvanced RA | 17643 | Wrist,MCPs | BME scores according to RAMRIS, evaluated by two readers | 0.99 and0.870.99 and0.94 | *Sens>95%Sens>95% |
↵* Sensitivity for the presence of BME (defined as BME score ≥1) with T2 sequence as reference.
MCP, metacarpophalangeal; nd,not determined; PIP, proximal interphalangeal; RA, rheumatoid arthritis; RAMRIS, RA MRI score; STIR, Short-TI Inversion Recovery.