Based on the results of stereotactic biopsy, we evaluated in a prospective fashion the efficiency of l-3-[123I]iodo-alpha-methyltyrosine-single-photon emission tomography (SPET) and [18F]fluorodeoxyglucose (FDG) positron emission tomography (PET) in the detection and grading of recurrences in patients previously treated for gliomas. The patient population comprised 30 individuals, nine with astrocytomas of grade II, ten with astrocytomas of grade IV, three with oligoastrocytomas of grade II, six with oligodendrogliomas of grade II and two with anaplastic oligodendrogliomas of grade III) suspected of recurrence and scheduled for further treatment. IMT SPET data were acquired using either by dual-or a triple-headed SPET camera, Multispect 2/3. FDG uptake was measured with an ECAT ART PET camera. Two independent observers classified PET and SPET images as positive or negative for tumour tissue. Uptake of FDG and IMT was evaluated visually and, in the case of IMT, also quantitatively by calculating the ratios between tracer accumulation in the lesion and the unaffected contralateral regions of reference using the region of interest (ROI) technique. The PET and SPET results were compared with the histopathological findings obtained either by stereotactic biopsy or in one case by open surgery. Glucose metabolism and amino acid uptake of recurrences of brain tumours as assessed by FDG-PET and IMT-SPET correlated highly with the histopathological findings. Based on the histopathological data, FDG-PET and IMT-SPET findings confirmed recurrence in all cases of high-grade gliomas (IV). A difference could be demonstrated in low-grade (II-III) tumour recurrences. True-positive IMT-SPET results were found in 86% of grade III and 75% of grade II recurrences, whereas FDG-PET yielded a sensitivity of 71% in tumours of grade III and 50% in those of grade II. With respect to the grade of malignancy of brain tumours at recurrence, IMT-SPET, in contrast to FDG-PET, does not permit adequate in vivo grading of non-mixed brain tumours of astrocytic or oligodendroglial origin. However, in this study FDG-PET did not permit discrimination between upgrading of low-grade oligoastrocytomas (II) into anaplastic oligodendrogliomas (III) and upgrading into glioblastomas (IV) The results of this study indicate that FDG-PET and IMT-SPET are equivalent to stereotactic biopsy in their ability to identify high-grade tumours at recurrence. IMT-SPET proved to be superior to FDG-PET in confirming low-grade recurrences. In the case of suspected progression of the grade of malignancy in ordinary gliomas, FDG-PET correlated significantly with the histopathological grading, whereas IMT-SPET did not. However, tumour grading by FDG-PET has a limitation in mixed brain tumours in that it is not possible to discriminate between progression of the oligo- versus the astrocytic tumour entity. In this case histopathological evaluation of the tumour grade remains necessary.