Aim: Assessment of the clinical value of (18)F-FDG-PET for detection of recurrent head and neck cancer, local lymph node involvement and distant metastases comparing a qualitative visual with a semiquantitative analysis (SUV values).
Patients, methods: Retrospective evaluation of 73 (18)F-FDG PET studies in 55 patients by use of a four-step qualitative visual grading system and calculation of standard uptake values in pathological lesions. Calculation of SUV values in normal regions for generating a map of physiological (18)F-FDG distribution. Correlation to histopathological findings and clinical follow-up.
Results: 1. Qualitative visual analysis of (18)F-FDG PET studies: a) local recurrence sensitivity 79%, specificity 97%, positive predictive value 95%, negative predictive value 85%, and diagnostic accuracy 89%; b) local metastatic lymph nodes 100%, 95%, 85%, 100%, 96%; c) distant metastases 100%, 98%, 86%, 100%, 98%, respectively. 2. Semiquantitative analysis had only little incremental, non-significant value in comparison to qualitative visual analysis for the detection of a local recurrence in two patients: a) local recurrence: sensitivity 83%, specificity 100%, positive predictive value 100%, negative predictive value 88%, and diagnostic accuracy 93%; b) local metastatic lymph nodes or c) distant metastases did not change in comparison to qualitative visual analysis.
Conclusion: (18)F-FDG PET is an effective tool for re-staging of patients with suspected recurrence after therapy for head and neck cancer.