Original Article
18F-FDG-PET/CT in the assessment of pulmonary solitary nodules: comparison of different analysis methods and risk variables in the prediction of malignancy
Abstract
Objective: To compare the diagnostic performance of different metabolical, morphological and clinical criteria for correct presurgical classification of the solitary pulmonary nodule (SPN).
Methods: Fifty-five patients, with SPN were retrospectively analyzed. All patients underwent preoperative 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography (CT). Maximum diameter in CT, maximum standard uptake value (SUVmax), histopathologic result, age, smoking history and gender were obtained. Different criteria were established to classify a SPN as malignant: (I) visually detectable metabolism, (II) SUVmax >2.5 regardless of SPN diameter, (III) SUVmax threshold depending of SPN diameter, and (IV) ratio SUVmax/diameter greater than 1. For each criterion, statistical diagnostic parameters were obtained. Receiver operating characteristic (ROC) analysis was performed to select the best diagnostic SUVmax and SUVmax/diameter cutoff. Additionally, a predictive model of malignancy of the SPN was derived by multivariate logistic regression.
Results: Fifteen SPN (27.3%) were benign and 40 (72.7%) malignant. The mean values ± standard deviation (SD) of SPN diameter and SUVmax were 1.93±0.57 cm and 3.93±2.67 respectively. Sensitivity (Se) and specificity (Sp) of the different diagnostic criteria were (I): 97.5% and 13.1%; (II) 67.5% and 53.3%; (III) 70% and 53.3%; and (IV) 85% and 33.3%, respectively. The SUVmax cut-off value with the best diagnostic performance was 1.95 (Se: 80%; Sp: 53.3%). The predictive model had a Se of 87.5% and Sp of 46.7%. The SUVmax was independent variables to predict malignancy.
Conclusions: The assessment by semiquantitative methods did not improve the Se of visual analysis. The limited Sp was independent on the method used. However, the predictive model combining SUVmax and age was the best diagnostic approach.
Methods: Fifty-five patients, with SPN were retrospectively analyzed. All patients underwent preoperative 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography (CT). Maximum diameter in CT, maximum standard uptake value (SUVmax), histopathologic result, age, smoking history and gender were obtained. Different criteria were established to classify a SPN as malignant: (I) visually detectable metabolism, (II) SUVmax >2.5 regardless of SPN diameter, (III) SUVmax threshold depending of SPN diameter, and (IV) ratio SUVmax/diameter greater than 1. For each criterion, statistical diagnostic parameters were obtained. Receiver operating characteristic (ROC) analysis was performed to select the best diagnostic SUVmax and SUVmax/diameter cutoff. Additionally, a predictive model of malignancy of the SPN was derived by multivariate logistic regression.
Results: Fifteen SPN (27.3%) were benign and 40 (72.7%) malignant. The mean values ± standard deviation (SD) of SPN diameter and SUVmax were 1.93±0.57 cm and 3.93±2.67 respectively. Sensitivity (Se) and specificity (Sp) of the different diagnostic criteria were (I): 97.5% and 13.1%; (II) 67.5% and 53.3%; (III) 70% and 53.3%; and (IV) 85% and 33.3%, respectively. The SUVmax cut-off value with the best diagnostic performance was 1.95 (Se: 80%; Sp: 53.3%). The predictive model had a Se of 87.5% and Sp of 46.7%. The SUVmax was independent variables to predict malignancy.
Conclusions: The assessment by semiquantitative methods did not improve the Se of visual analysis. The limited Sp was independent on the method used. However, the predictive model combining SUVmax and age was the best diagnostic approach.
