Analysis of discordant PET and CT findings in 18F-FDG PET-CT scans in the management of oncology patients

OBJECTIVES: Discordant findings are often noted between PET-CT and CT images of 18F-FDG PET-CT scans. This study aimed at determining the significance of these findings in the management of oncology patients. METHODS: This was an observational, descriptive study. Hence retrospective analysis of all discordant findings in oncology patients undergoing a PETCT imaging between Jan 2013 - Jan 2016 was done. Those patients who had a follow up period of minimum one year in either of the following forms - repeat PETCT imaging, other radiological imaging, clinical or histopathological evidence were included. From all the discordant lesions, the sensitivity, specificity, positive predictive, negative predictive value and accuracy of both PET-CT and CT modalities were determined. RESULTS: Of 348 discordant lesions, 16.7% was noted in soft tissues, 25% in viscera, 28.7% in lungs, 14.1% in lymph nodes and 15.5% in bones. At the end of follow up, 15.2% lesions were PET true positive, 57.5% PET true negative, 10.1% CT true positive lesions, 13.8% CT true negative and 3.4% were inconclusive. 18F-FDG PET-CT is superior to CT imaging in oncology. CONCLUSION: From our study it was found that overall, 57.5% of the discordant lesions were false positive on CT (true negative on PET) and 10.1% were true positive on CT (False negative on PET). The high proportion of CT false positive findings suggests that suspicious lesions on CT with no PET uptake in the setting of a malignancy are unlikely to be significant. However the small percentage of false negatives on PET warrants cautious interpretation of findings. A thorough history, examination, knowledge of the biopsy of the malignancy and treatment history is imperative to their accurate interpretation. It was also noted that overall, 13.8% of the discordant lesions were false positive on PET (true negative on CT). This has led to unwarranted biopsies and invasive procedures, yet the presence of increased abnormal FDG activity cannot be dismissed as benign. The most common cause of false positive uptake is inflammatory changes which is a drawback of the tracer used. It was noted that 15.2% of the discordant lesions were true positive on PET-CT (false negative on CT). This emphasises the fact that 18F-FDG PET-CT plays a major role in staging the disease and alters treatment decisions. In summary, the overall sensitivity of PET-CT was 60.2%, while that of CT was 39.7%, specificity of PET-CT and CT was 80.6% and 20.4% respectively. The positive predictive value for PET-CT was 52.4% as opposed to 47.6% for CT, the negative predictive value for PET-CT was 85.1% compared to 14.9% for CT and the accuracy of PET-CT was 75.3% and that of CT was 24.7%. Considering the site of the discordant lesion, positive findings on PET-CT in the viscera and soft tissue have low reliability in view of the high false positive findings. The absence of FDG activity in a discordant lesion in lungs is unlikely to be malignant in view of the high negative predictive value. A positive finding on PET-CT in a discordant lesion detected in the bones/ bone marrow and lymph nodes has a higher probability of being malignant even though CT is reported as normal. The comparison made between PET-CT and CT imaging highlighted that PET-CT had a superior advantage over CT alone in the diagnostic imaging of Oncology patients and should be considered as the first-line imaging modality