Integrated imaging in peripheral nerve lesions in type 1 neurofibromatosis

Our aim was to evaluate the role of Ultrasonography (US) and Magnetic Resonance (MR) in the staging and follow-up of peripheral nerve lesions in type 1 Neurofibromatosis (NF1).

We evaluated forty-six NF1 patients (16 males and 30 females, age range 12-65 years, mean age 34 years) affected by 51 soft tissue lesions, clinically diagnosed as subcutaneous (18 cases) and plexiform (33 cases) neurofibromas. The lesions were studied to identify site, size, extension, relationship with surrounding structures, and any features indicating malignant transformation. All patients underwent at least one US examination performed with a 7.5 - 10 MHz linear multi-frequency probe. All subcutaneous neurofibromas larger than 2 cm, all superficial or deep plexiform neurofibromas not clearly defined at US, and all large plexiform neurofibromas with massive involvement of surrounding tissues were studied by MR using 0.2 T magnet (Artoscan) or 1.5 T magnet (Magnetom Vision Plus Siemens). Fifteen patients underwent surgery; the remaining cases (31) were followed up according to the National Institutes of Health (NIH) protocol.

The subcutaneous neurofibromas smaller than 2 cm showed homogeneous hypoechoic echotexture (10 cases), whereas lesions equal to or larger than 2 cm showed a characteristic target pattern (peripheral hypoechoic rim with hyperechoic core). All lesions larger than 2 cm studied by MRI displayed the target pattern in T2-weighted sequences, with peripheral signal hyperintensity and central signal hypointensity. At US, the 33 plexiform neurofibromas had an irregular shape with undefined borders; the patterns were classified as follows: 1) superficial plate-like lesions (14 cases); 2) superficial lesions with deep digitations (8 cases); 3) deep multi-lobulated lesions (5 cases); 4) extensive lesions massively involving soft tissues (6 cases). MRI was performed in 11 cases, showing an isointense or moderately hyperintense signal on T1-weighted sequences, high signal intensity on T2-weighted sequences, and, after intravenous gadolinium administration, variable enhancement on T1-weighted sequences.

The results obtained in subcutaneous neurofibromas indicate a clear relationship between lesion size and US-MR imaging, with similar local staging. Since MRI failed to provide additional information, US may be considered the first-choice, and sometimes decisive, examination in these cases. As regards plexiform neurofibromas, US only proved useful in staging superficial lesions and partly deep multi-lobulated lesions as it accurately depicted lesion size, but not local extension; MRI is therefore useful in the preoperative staging of lesions. In extensive lesions, US proved unable to provide accurate and complete local staging of the lesions, so that MRI should be systematically used in these cases.