Abstract
217
Objectives: The standardized uptake value (SUV) of physiologic F-18 fluorodeoxyglucose (FDG) liver uptake sometimes varies greatly from individual to individual, in the absence of obvious pathology in the liver. Previous studies have reported that body weight (BW) and body mass index (BMI) were significant factors associated with physiologic FDG uptake of the liver. On the other hands, the clinical significance of increased or decreased liver uptake on FDG-positron emission tomography (PET) is not well understood. In daily clinical practice, reduction of physiologic FDG uptake in the liver is sometimes identified on FDG-PET in malnourished patients, and we were interested in the relationship between decreased hepatic FDG uptake and patients with malnutrition and/or cachexia. The purpose of this retrospective study was to investigate clinical characteristics of patients with diffusely increased and decreased physiologic FDG uptake in the liver on FDG-PET, including whether or not diffusely decreased FDG uptake in the liver was associated with cachexia. We investigated clinical characteristics of patients with extremely increased or decreased physiologic FDG uptake of the liver and their prognosis.
Methods: 1487 PET/computed tomography (CT) scans of patients with known or suspected malignancy between 2015 and 2016 were retrospectively analyzed. A spherical VOI (3 cm in diameter) was set on the right lobe of the liver to calculate the SUVmean. A reference range of liver SUVmean was established from the 2.5th (lower limit) to the 97.5th (upper limit) percentiles of measurements in the 1487 PET/CT scans. Two outlier groups were defined: an extremely high liver uptake group of patients with liver SUVmean > 97.5th percentile (HG) and an extremely low liver uptake group of patients with liver SUVmean < 2.5th percentile (LG), respectively. A control group of cases whose FDG uptake in the liver was considered to be normal was also defined. Physical and laboratory data among a control group (n = 30), the HG (n = 36), and the LG (n = 36) were compared. Overall survival (OS) of the three groups was also compared (Suppl. 1).
Results: The average liver SUVmean of all scans (n = 1487) was 2.36 ± 0.32. The average liver SUVmean of controls was 2.41 ± 0.28, not significantly different from that of all patients. The reference range of liver SUVmean was set to 1.79-3.03 (2.5th -97.5th percentiles). Then, the liver FDG uptakes of the HG and LG were defined as SUVmean ≥ 3.04 and SUVmean ≤ 1.78, respectively (Fig. 1). Suppl. 2 shows representative PET images of extremely low (a, 47-year-old female after total gastrectomy for gastric cancer), normal (b, 60-year-old male with suspected lung cancer), and extremely high (c, 41-year-old-female with cervical cancer recurrence) liver FDG uptake in the absence of liver tumors. BW and BMI in the HG (SUVmean ≥ 3.04) were significantly higher than those in the control group (Fig. 2). The LG cases (SUVmean ≤ 1.78) had anemia (lower red blood cell count and hemoglobin), impaired liver function (lower total protein, albumin, and cholinesterase), and systemic inflammation (higher CRP) (Fig. 3). They were also in a poor nutritional state. The characteristics of LG cases had many things in common with those of cachectic patients. Indeed, when referring to the definition of cachexia proposed by ESPEN (the European Society for Clinical Nutrition and Metabolism) Consensus Statement, 36.1% of LG cases met the diagnostic criteria for cachexia. Moreover, in LG cases with viable and/or recurrent malignant lesions on FDG-PET, the proportion of cachexia increased by 52.6% (Fig. 4). The OS of LG cases (median 33 months) was significantly worse than that of controls and HG cases (Fig. 5).
Conclusions: Our data indicated that cancer patients with extremely decreased liver FDG uptake were likely to have cancer cachexia and a lower OS.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.