The majority of abnormal findings or lesions on T2-weighted fast spin-echo (FSE) magnetic resonance imaging (MRI) are hyperintense due to increased perfusion or fluid content, such as infections, tumours or synovitis. Hypointense lesions on T2-weighted images (both fat-suppressed and non-fat-suppressed) are less common and can sometimes be overlooked. Such lesions have limited differential diagnostic possibilities, and include vacuum phenomenon, loose body, tenosynovial giant cell tumour, rheumatoid arthritis, haemochromatosis, gout, amyloid, chondrocalcinosis, hydroxyapetite deposition disease, lipoma arborescens, arthrofibrosis and iatrogenic lesions. These lesions often show characteristic appearances and predilections in the knee. In this article, the authors describe the MRI features of hypointense T2 lesions on routine knee MRI and outline a systematic diagnostic approach towards their evaluation. Key Points • Hypointense lesions on T2 images (T2 Dark Lesions) encompass limited diagnostic possibilities. • T2 Dark lesions often show characteristic appearances and predilections in the knee. • A systematic diagnostic approach will help radiologists make the correct diagnosis.

No abstract

Objectives: Scoliosis is a common orthopedic problem in patients with neurofibromatosis 1 (NF1). Spinal deformities are found in 77% of all NF1 cases, with no widely accepted etiology. This study aimed to evaluate the frequency and types of scoliosis in NF1 patients using whole-body magnetic resonance imaging and to assess the association of intraspinal and paraspinal tumors with the imaging findings of scoliosis.Methods: A total of 122 NF1 patients with whole-body magnetic resonance imaging were found from the electronic medical records. Ninetyseven cases that met the inclusion criteria were identified. All patients underwent 3-T magnetic resonance imaging with automated software fusion of the 3 sets of short TI inversion recovery and 3-dimensional T1-weighted coronal images. Frequency and location of scoliosis and intraspinal and paraspinal tumors were recorded. Patients with severe dystrophic-type scoliosis were separately identified, and Cobb angles were measured for all such cases. Association analysis was performed. A P value less than 0.05 was considered statistically significant.Results: Ninety-seven patients with NF1 were evaluated. Two had prior spinal surgery and were excluded. The final sample of 95 patients included 33 (35%) men and 62 (65%) women with a mean ± SD body mass index of 25.82 (4.96) kg/m 2 . Of the 95 patients, 43 (45.3%) had scoliosis, 13 of 43 (30.2%) of which were severely angled. Of the 95 patients, 25 (26.3%) had locoregional tumor presence. Intraclass correlation for Cobb angles measured 0.99 (confidence interval, 0.98-1.0). Fisher exact test determined no association between scoliosis and presence of either paraspinal or intraspinal tumors ( P = 0.485). There was also no association between the tumors and severe dystrophic scoliosis ( P = 1.0). Conclusions:This study found no association between the presence of locoregional spinal tumors and scoliosis in NF1 patients. This work adds to the body of knowledge of scoliosis in NF1 patients and infers that presence of scoliosis should not mandate immediate search for locoregional spinal tumors.

No abstract