In suspected scaphoid fracture the initial scaphoid series plain radiographs are 84-94% sensitive for scaphoid fractures. Patients are immobilised awaiting diagnosis. Unnecessary lengthy immobilisation leads to lost productivity and may leave the wrist stiff. Early accurate diagnosis would improve patient management. Although Magnetic Resonance Imaging (MRI) has come to be regarded as the gold standard in identifying occult scaphoid injury, recent evidence suggests Computer Tomography (CT) to be more accurate in identifying scaphoid cortical fracture. Additionally CT and USS are frequently a more available resource than MRI. We hypothesised that 16 slice CT is superior to high spatial resolution Ultrasonography (USS) in the diagnosis of radiograph negative suspected cortical scaphoid fracture and that a 5 point clinical examination will help to identify patients most likely to have sustained a fracture within this group. 100 patients with two negative scaphoid series and at least two out of five established clinical signs of scaphoid injury (anatomical snuffbox tenderness (AST), scaphoid tubercle tenderness (STT), effusion, pain on circumduction and pain on axial loading) were prospectively investigated with CT and USS. MRI was arranged for patient with persistent symptoms but negative CT/USS.Background
Methods
Scaphoid fracture is the most common undiagnosed fracture. Occult scaphoid fractures occur in 20-25 percent of cases where the initial X-rays are negative. Currently, there is no consensus as to the most appropriate investigation to diagnose these occult frctures. At our institution MRI has been used for this purpose for over 3 years. We report on our experience and discuss the results. All patients with occult scaphoid fractures who underwent MRI scans over a 3 year period were included in the study. There was a total of 619 patients. From the original cohort 611 (98.7%) agreed to have a scan, 6 (0.97%) were claustrophobic and did not undergo the investigation and 2 (0.34%) refused an examination. 86 percent of the cases were less than 30 years of age. Imaging was performed on a one Tiesla Siemen's scanner using a dedicated wrist coil. Coronal 3mm T1 and STIR images were obtained using a 12cm field of view as standard. Average scanning time was 7 minutes.Introduction
Materials and methods
