Optimizing ¹⁸F-FDG Uptake Time Before Imaging Improves the Accuracy of PET/CT in Liver Lesions

Abstract

Beaumont Health System, Royal Oak, Michigan 18 F-FDG PET/CT has emerged as one of the fastest-growing imaging modalities. A shorter protocol results in a lower targetto-background ratio, which can increase the challenge of identifying mildly 18 F-FDG-avid lesions and differentiating inflammatory or physiologic activity from malignant activity. The purpose of this study was to determine the delay between radiotracer injection and imaging that optimizes target-to-background ratio while maintaining counts high enough to ensure scan sensitivity. Methods: The study included 140 patients (66 male and 74 female; age range, 42-95 y) with suspected hepatic lesions as seen on an 18 F-FDG PET scan. SUV was determined as regionof-interest activity/(dose/total body weight). Results: The mean injected dose was 610 ± 66.6 MBq (16.5 ±1.8 mCi), with a mean glucose level of 107 ± 26.6 mg/dL (standardized to 90 mg/dL). The uptake time before imaging ranged from 61 to 158 min, with a mean of 108.8 ± 24.8 min. The P values for the correlation of SUV to time were 0.004, 0.003, and 0.0001 for malignant lesions, benign lesions, and background hepatic tissue, respectively. Conclusion: An approximately 90-min time window from 18 F-FDG injection to PET imaging would significantly improve target-to-background ratio and, thus, quantitation and visual interpretation. This benefit outweighs the minimal loss in patient throughput. Int egrated 18 F-FDG PET/CT has emerged as one of the fastest-growing imaging modalities. The recent expansion of coverage from the Centers for Medicare and Medicaid Services has made its use even more feasible for communitybased hospitals and independent imaging centers. To maximize throughput, many of these centers use a 50-to 60-min uptake time after 18 F-FDG injection while arguing that the counting statistics are improved using this protocol (1-3). However, this shorter protocol results in a lower targetto-background ratio, which can increase the challenge of identifying mildly 18 F-FDG-avid lesions and differentiating inflammatory or physiologic activity from malignant activity (4-6).The purpose of this study was to determine the delay between radiotracer injection and imaging that optimizes target-to-background ratio while maintaining counts high enough to ensure scan sensitivity. Hepatic lesions and background hepatic tissue were chosen as the test and control tissues because 18 F-FDG has a relatively predictable parabolic uptake pattern in the liver (7,8). Additionally, the heterogeneous or mottled appearance of the liver often presents a challenge during image interpretation by producing areas of benign focal uptake or masking uptake within a malignant lesion (6). MATERIALS AND METHODSThe study was approved by the human investigation committee at William Beaumont Hospital (study 2012-226). The population comprised 140 patients (66 male and 74 female; age range, 42-95 y) with suspected hepatic lesions as seen on an 18 F-FDG PET scan. All enrolled patients had a primary malignancy and either had been referred...

Image features for the future in stage IV non-small cell lung cancer

“…Protocol deviations are quite common in clinical practice, a written survey among all Austrian PET/CT centers, showed that the average uptake time there was 55 minutes, with a range of 45 to 75 minutes, which is lower than in the EANM guidelines [65]. The optimal uptake time is as well tumor site specific, for liver imaging using 18 F-FDG PET the optimal uptake time was shown to be 90 minutes after FDG injection [66]. A possible reason for deviations in our study is that the radiopharmaceutical start time, extracted from the Digital Imaging and Communications in Medicine (DICOM) header, was not the time of FDG administration, as we assumed.…”

“…Also, when comparing an exposure of 60 mA with 80 mA on the same scanner, or scanner 'B' with 'O', no trend indicating a dependency on bin width could be observed. In this case, the median number of stable features was 66.5 (55-70) and 67.5 (56)(57)(58)(59)(60)(61)(62)(63)(64)(65)(66)(67)(68)(69)(70)(71)(72), respectively. Note that the total number of stable features was much lower when comparing different slice thicknesses.…”