Hybrid peripheral nerve sheath tumours: MRI features with pathological correlation in 24 cases

Le résumé fourni par la source

OBJECTIVES: To describe the MRI features of histologically proven hybrid peripheral nerve sheath tumours (HPNST). METHODS: A retrospective analysis of the MRI features of 24 histologically proven cases of HPNST over 7 years. Demographic data obtained from clinical records included age, gender, and date of diagnosis. Two readers independently assessed MRI studies and assessed the following features: involvement of a major nerve, intramuscular location, lesion morphology, entering nerve sign, exiting nerve sign, target sign, fascicular sign, split fat sign, and ancient change (cystic change). Inter-observer agreement was assessed with Cohen's kappa coefficient. Histological diagnosis was based on either image-guided needle biopsy or resection histology. RESULTS: The study included 9 males and 15 females with mean age 50 years (range: 24-78 years). Nine tumours (35%) involved a major nerve including spinal roots (5), radial (1), median (1), tibial (1), and axillary (1), while 5 (21%) tumours were intramuscular. The mean tumour size was 4.2 cm (standard deviation of 2.4 cm). The frequency of MRI features was as follows: lobular contour (71%; 17/24), ancient change (38%; 9/24), fascicular sign (17%; 4/24), entering nerve sign (21%; 5/24), exiting nerve sign (13%; 3/24), target sign (13%; 3/24), and split fat sign (8%; 2/24). Inter-observer agreement was high, ranging from 0.7 to 0.83. CONCLUSIONS: HPNST infrequently demonstrate the classical MRI signs of benign peripheral nerve sheath tumours, but commonly have a lobular morphology and can show ancient/cystic change. ADVANCES IN KNOWLEDGE: This is the first study in the literature analysing the MRI features of histologically proven HPNST. HPNST infrequently shows the classical MRI signs that would be expected with benign peripheral nerve sheath tumours although commonly have a lobular morphology and show cystic change.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.