Oncologic PET/CT interpretation and reporting approaches. Survey in clinical practice.

AIM: To elucidate techniques most commonly used for interpreting oncologic PET/CT studies. This survey forms a basis to work on standardization of reporting and highlight the most important issues to be addressed. METHODS: A web-based survey of 329 PET/CT imaging specialists was designed with the intent to determine image interpretation patterns. The questionnaire consisted of 19 questions. Of the 329 participants, 230 were nuclear medicine specialists, 46 were radiologists, and 53 had dual-board certification. RESULTS: Report ofstandardized uptake values (SUV) is not consistent;only50.2% of respondents always report SUVs, while 45.2% report only if needed or requested. 80.9% of respondents indicated that reporting of SUV is only appropriate when its limitations are understood whereby a large majority prefer to report SUVmax. Maximum intensity projection (MIP) images are almost always reviewed by 91.1% of the respondents. An accurate and detailed clinical history is considered an essential element for reading PET/CT studies by 84.0%, but only 20.7% report that this is always available. The most common self-reported average time for reviewing and reporting of whole body PET/CT (with no prior comparison scan) was 15-20 min (27.5%). CONCLUSION: PET readers have considerable reservations regarding the use and reporting of SUVs. SUVmax is more frequently used than SUVmean. Evaluation of MIP images is considered an important element of PET/CT interpretation. Although availability of sufficient patient's history is considered essential, this is rarely available.

Performance of 2-deoxy-2-[F-18]fluoro-D-glucose positron emission tomography and integrated PET/CT in restaged breast cancer patients.

PURPOSE: This study was conducted to compare the clinical stage derived from 2-deoxy-2-[F-18]fluoro-D-glucose (FDG) positron emission tomography (PET) to that of integrated PET/computed tomography (CT) in restaged breast cancer patients. PROCEDURES: Fifty-eight female patients (age range 29-80 years, mean age +/-SD, 53.3 +/- 11.7 years) underwent PET/CT restaging for breast cancer. Two experienced nuclear medicine physicians interpreted PET images. A radiologist was added for reading PET/CT studies. A patient-based analysis was performed. Histopathological findings, correlative imaging studies, changes in number, size, and hypermetabolic activity of suspicious lesions and/or patient outcome served as standard of reference for determining the diagnostic accuracy of both modalities. RESULTS: PET staged 79.3% (46/58) of the patients correctly, overstaged seven (12.1%), and understaged five patients (8.6%). Integrated PET/CT staged 89.7% (52/58) of the patients correctly, overstaged four (6.9%), and understaged two patients (3.4%). The staging accuracy of PET/CT was not significantly better than that of PET alone (p = 0.059). Lesions exhibiting mild hypermetabolic activity, benign inflammatory lesions, and physiological variants largely explained incorrect PET findings. CONCLUSION: Integrated PET/CT only marginally improves the restaging accuracy over PET alone (p = 0.059) in breast cancer patients.