Fluorine-18 fluorodeoxyglucose (FDG)-positron emission tomography (PET) is useful in Hodgkin and B-cell lymphomas. Few data exist on T-cell and natural killer (NK)-cell lymphomas. Thirty consecutive T-cell and NK-cell lymphomas were investigated with PET-computerized tomography (CT). In 12 NK-cell lymphomas, all nasal/extranasal lesions were FDG-avid. In nasal/maxillary areas, FDG-avid tumours were consistently more localised than on CT, suggesting that soft tissue masses on CT were partly due to inflammation. These findings have important implications in radiotherapy planning. In two NK-cell lymphomas, PET did not detect morphologically occult marrow infiltration uncovered by in situ hybridisation for Epstein-Barr-virus-encoded small RNA. In angioimmunoblastic lymphoma (n = 7), peripheral T-cell lymphoma, unspecified (PTCL-U, n = 4) and anaplastic large cell lymphoma (ALCL, n = 3), involved nodal/extranodal sites shown on CT and/or biopsy were concordantly PET-positive. In one PTCL-U, PET detected FDG-avid marrow infiltrations not shown on biopsies. In contrast, cutaneous ALCL (n = 1) and mycosis fungoides (n = 2) showed minimal FDG uptake. In one case of T-cell large granular lymphocyte leukaemia, marrow, nodal and bowel infiltrations were not FDG-avid. PET maximum standardised uptake value did not correlate with clinicopathological features and prognosis. These observations defined the pre-treatment value of PET-CT in T-cell and NK-cell lymphomas. The post-treatment role requires further studies.