Abstract
Goal: The purpose of this study was to demonstrate that standard uptake value (SUV) calculations standardized to normal liver parenchyma, and utilizing previously defined thresholds for differentiating benign and malignant disease, yield results superior to non-standardized SUVs. This methodology can be applied to all fluorodeoxyglucose positron emission tomography (FDG PET) interpretations.
Methods: Patients (n=132) with a history of primary lung carcinoma were studied with FDG PET to determine the presence or absence of metastatic disease. Images were assessed using semi-quantitative analysis with SUVmax generation. All SUVs were calculated with and without individualized correction for normal liver parenchymal uptake and the standard uptake value optimized (SUVopt) was derived. Results of corrected and un-corrected SUVs were compared with the pathologic results in two clinical disease categories: 1. pleural metastases (n= 87), and 2. adrenal metastases (n= 45).
Results: The sensitivity and specificity for corrected versus uncorrected SUVs for pleural metastases were: 95.6% versus 93.8%, and 77.0% versus 42.1%, respectively. For adrenal metastases the results were: 94.5% versus 91.6%, and 88.9% versus 45.0%, respectively.
Conclusions: In patients with primary lung cancer, the utilization of SUVs corrected for the normal liver parenchyma provide more accurate results in defining the presence or absence of metastatic disease to pleura and adrenal glands.