OBJECTIVE: The purpose of our study was to evaluate a visual PET/CT scoring system for the differentiation of benign and early malignant esophageal uptake. MATERIALS AND METHODS: Thirty-six consecutive patients with precancerous or early malignant esophageal lesions including Barrett's esophagus, Tis, T1, and T2 adenocarcinomas were eligible. Findings of these patients were compared with 66 patients who had reported increased esophageal (18)F-FDG uptake due to benign esophageal disorders. Lesions were evaluated with scores using the following characteristics in PET/CT: FDG uptake intensity (low = 0, moderate = 1, high = 2), FDG uptake eccentricity (concentric = 0, eccentric = 1), FDG uptake focality (diffuse = 0, segmental = 1, focal = 2), esophageal thickness on the CT component (normal = 0, thickening = 1, mass = 2), and location (distal third of the esophagus = 0, middle third of the esophagus = 1, proximal third of the esophagus = 2). RESULTS: Early malignant lesions had higher scores in FDG uptake intensity (p = 0.003; chi-square), eccentricity (p < 0.001), and focality (p < 0.001) compared with benign lesions. No significant difference was seen in esophageal thickness on CT (p = 0.168) and in location of the lesion (p = 0.291). Binary logistic regression analysis with a stepwise forward inclusion of all score components including the maximum standardized uptake value (SUV) of the lesions revealed that a total score combining eccentricity and focality scores has the highest accuracy of predicting early malignant disease. Using a threshold of equal or higher than 2 in the combined total focality-eccentricity score, the sensitivity was 83.3% and specificity was 68.2% for predicting early malignant disease. CONCLUSION: Focality and eccentricity of FDG uptake prove to be valuable PET/CT characteristics for the differentiation of nonspecific FDG uptake in the esophagus.
OBJECTIVE: The purpose of our study was to evaluate a visual PET/CT scoring system for the differentiation of benign and early malignant esophageal uptake. MATERIALS AND METHODS: Thirty-six consecutive patients with precancerous or early malignant esophageal lesions including Barrett's esophagus, Tis, T1, and T2 adenocarcinomas were eligible. Findings of these patients were compared with 66 patients who had reported increased esophageal (18)F-FDG uptake due to benign esophageal disorders. Lesions were evaluated with scores using the following characteristics in PET/CT: FDG uptake intensity (low = 0, moderate = 1, high = 2), FDG uptake eccentricity (concentric = 0, eccentric = 1), FDG uptake focality (diffuse = 0, segmental = 1, focal = 2), esophageal thickness on the CT component (normal = 0, thickening = 1, mass = 2), and location (distal third of the esophagus = 0, middle third of the esophagus = 1, proximal third of the esophagus = 2). RESULTS: Early malignant lesions had higher scores in FDG uptake intensity (p = 0.003; chi-square), eccentricity (p < 0.001), and focality (p < 0.001) compared with benign lesions. No significant difference was seen in esophageal thickness on CT (p = 0.168) and in location of the lesion (p = 0.291). Binary logistic regression analysis with a stepwise forward inclusion of all score components including the maximum standardized uptake value (SUV) of the lesions revealed that a total score combining eccentricity and focality scores has the highest accuracy of predicting early malignant disease. Using a threshold of equal or higher than 2 in the combined total focality-eccentricity score, the sensitivity was 83.3% and specificity was 68.2% for predicting early malignant disease. CONCLUSION: Focality and eccentricity of FDG uptake prove to be valuable PET/CT characteristics for the differentiation of nonspecific FDG uptake in the esophagus.