Magnetic resonance (MR) imaging was performed in 116 patients in whom a parotid mass lesion was clinically suspected. Eighty-six patients had benign disease. The 30 patients in whom a malignant tumor was found were further evaluated. To determine which features are characteristic of malignant parotid tumors, spin-echo T1- and T2-weighted images of malignant lesions in the parotid gland were compared with those of benign disease. In our series, tumor margins, homogeneity, or signal intensity were not discriminative factors to correctly predict benign or malignant disease. Infiltration into deep structures (eg, the parapharyngeal space, muscles, and bone) was observed only in malignant tumors. Infiltration into subcutaneous fat was noticed in malignant as well as in inflammatory disease. No statistically significant correlation was found between tumor grade and MR imaging features in malignant disease. MR imaging is useful in delineating malignant tumors but is unreliable in correctly predicting the histologic nature of a mass lesion in the parotid gland.

In a retrospective study of 73 patients operated on for cerebellar pilocytic astrocytomas, results of treatment, outcome and biological behaviour of residual tumour were analysed. Complete tumour resection proven by CT or MRI scans within 1 year after surgery was achieved only in 69% of cases. In 31% of cases the surgeon's opinion on the extent of surgical resection was not borne out by the result of postoperative neuroimaging. Progression of residual tumour or tumour recurrence appeared in 19% of patients. 1 patient showed metastatic spread along the craniospinal axis, and in 1 patient malignant degeneration appeared during follow-up. Stable residual tumour or regression of residual tumour was seen in 14% of patients. Outcome after surgical treatment, which was combined with irradiation in 10 patients (14%), was favourable in 80% and unfavourable in 20% of patients. This outcome of treatment was not influenced by a second operation for progression of residual tumour or recurrent tumour. Characteristics of patients with tumour progression after the first operation did not differ from those of the whole group. There were 17 reoperations for residual or recurrent tumour, 10 of which took place within 4 years after the initial surgical treatment. Surgery-related morbidity was 15% and mortality 4%. Irradiation to residual tumour in 8 patients was followed by complete regression in 1 patient, progression in 4 patients and no changes in 1 patient. For the remaining 2 patients the effect of irradiation on the residual tumour is unknown. Factors that determine the prognosis are discussed on the basis of this retrospective analysis and the data from the literature. It is concluded that optimal treatment for a cerebellar pilocytic astrocytoma does not consist solely in surgery with the aim of total tumour removal and careful tumour handling in order to avoid spread of tumour cells and subsequent metastases and additional radiation therapy is strictly selected cases, but also in posttreatment follow-up based on direct postoperative neuroimaging, preferably by MRI. An algorithm for postoperative follow-up management is presented.

No abstract

No abstract