6.7. Role of radiologic imaging in the study and diagnosis of NF1

NF1 is a genetic disorder that is still primarily diagnosed on the basis of its clinical features by using the widely accepted diagnostic NIH criteria, as in all cases of the present study . A diagnosis of NF1 in a patient always signifies a risk of having CNS manifestations at the time of diagnosis already or developing them at a later age. In children, the main diagnostic concern is the risk for optic gliomas.

The present study confirms that NF1 is, both clinically and radiologically, a highly heterogeneous disorder. Pathologic lesions of various types can be found practically anywhere in the CNS, either in the nervous tissue itself or in the neighbouring structures. Radiologic imaging can often detect the cause for the CNS-related symptoms, but it also reveals unexpected lesions, whose significance may need to be evaluated by further studies.

Although both CT and MRI have been established as valuable tools in detecting CNS manifestations, MRI is the method of choice for studying T2 hyperintense lesions of the brain, the intracranial extent of optic gliomas and lesions of the spinal cord and nerves and for differentiating optic nerve tumours from hydrops of the optic sheath. Because of a lack of ionized radiation, MRI is also suitable for repeated imaging when needed. The use and timing of CNS imaging in the diagnosis, screening and follow-up of NF1 patients has, however, remained somewhat controversial, because the clinical criteria of the disease are sufficient for the diagnosis of NF1 in most patients, and the most common manifestations of the disease, namely optic gliomas and T2 hyperintense lesions of the brain, have a benign course in most cases. Listernick et al. (1997), who considered regular clinical and neuro-ophthalmologic examination of symptomless optic gliomas necessary, suggested imaging only when symptoms occur. On the other hand, T2 hyperintense lesions have even been recommended as one of the criteria of NF1 in cases where the clinical criteria have not been met. In the study of Menor et al. (1998), MRI contributed to a definitive diagnosis of NF1 in 53% of the suspected cases and was recommended to confirm the diagnosis in young patients who did not fulfil the NIH criteria. The present study has also shown the significance of T2 hyperintense lesions in NF1 in children, in many of whom they confirmed the diagnosis. NF1 patients have an increased risk for brain and other malignancies, and malignant evolution of T2 hyperintense brain lesions on other than typical sites does occur. Screening studies have revealed specific variants of NF1, such as FSNF in Northern Finnish patients, and families with several members affected by brain tumours have been described, such as the mother and her son with brain astrocytomas. There is a trend to use MRI imaging more extensively and at an earlier age in the evaluation of NF1 patients, but CNS imaging should be considered on an individual basis by the multidisciplinary team which is responsible for the clinical follow-up of the patient.

The use of intravenous contrast medium has been recommended by several authors (Bonawitz et al. 1998) both in MRI and CT. In the present study, contrast medium was used when necessary in CT and MRI for diagnostic purposes, but not in patients screened for CNS lesions. In the latter group, only rare patients needed MRI scanning with contrast enhancement in order to obtain further information.

In general, the role of CT has diminished because of the risks caused by ionized radiation especially to the lens in orbital studies, but also because of the need for repeated follow-up examinations. The availability of MRI is, however, still limited, and the examination is time-consuming and expensive. The risks related to anesthesia and the use of contrast medium also need to be considered in the case of both CT and MRI.

The present study reflects the developing diagnostic and follow-up practices of NF1 in a modern university hospital setting. The new methods being developed, such as MR spectroscopic and diffusion studies and PET, have already shown promising results in the differentiation between benign and malignant or potentially malignant lesions in NF1.