A preliminary study on the impact of fever on the biodistribution of 18F-FDG in vivo

Objectives: Elevating body temperature may alter the biodistribution of ¹⁸F-FDG owing to affecting the glucose metabolism in various organs, and may lead to inaccurate findings and possible misinterpretation of ¹⁸F-FDG PET/CT images. In this study,we explore the effects of fever upon the biodistribution of ¹⁸F-FDG through evaluating the difference in semi-quantitative standardized uptake value (SUV) measurements of organs and tissues between febrile patients and patients with the normal body temperature.

Methods: ¹⁸F-FDG PET/CT in our department from January 2016 to June 2017 were retrospectively reviewed. A total of 69 patients with body temperature higher than 37.3℃ were selected as the experimental group and 69 patients with normal body temperature were selected as controls, which matched the experimental group in age, weight and sex during the same period. SUVmax in brain, thyroid, mediastinal blood pool, liver, spleen, bilateral adrenal glands, bone marrow, fat and muscle were measured. Age, blood glucose before examination, body temperature on the day of examination and laboratory test including red blood cells, hemoglobin, white blood cells, platelets, alanine aminotransferase, aspartate aminotransferase , total protein, estimated glomerular filtration rate(eGFR),triglyceride, C-reactive protein (CRP), thyroid stimulating hormone (TSH), free triiodothyronine (FT3) and free thyroxine (FT4) were collected. The SUVmaxs were compared between the experimental and the control group. Multi-factor analysis was performed if there are significant differences in SUVmax between the two groups.

Results: Compared with the control group, cerebral SUVmax [9.26 (6.97-11.4) vs 14.26 (12.81-16.39)], thyroid SUVmax [1.13 (0.92-1.35) vs 1.27 (1.06-1.56)], SUVmax in mediastinal blood pool [1.59 (1.43-1.73) vs 1.89 (1.78-2.04)], hepatic SUVmax (2.43±0.35 vs 2.74±0.31) of the experimental group significantly decreased (P<0.01), whereas adipose SUVmax (0.71±0.16 vs 0.64±0.13), splenic SUVmax [2.94 (2.42-3.64) vs 2.24 (2.02-2.41)], bone medullary SUVmax [3.11 (2.76-3.58) vs 2.20 (2.02-2.47)] of experimental group increased significantly (P<0.01). In the experimental group, all organs but the bone marrow did not show SUVmax statistically difference between the inflammation group and the tumor group. SUVmax of all organ did not differ between patients with FDG stress and patients without FDG stress. Univariate analysis demonstrated that body temperature on the day of examination, age, erythrocyte, hemoglobin, and eGFR were correlated with SUVmax in each organ. Multivariate analysis suggests that body temperature on the day of examination was independent factor of SUVmax in brain, thyroid, adipose, mediastinal blood pool, spleen, and bone marrow(P<0.01). SUVmaxs in Brain (r=-0.592), thyroid(r=-0.236) and the mediastinal blood pool(r=0.500) were negatively correlated with body temperature on the day of examination, whereas adipose SUVmax (r=0.275), splenic SUVmax (r=0.581), and the bone medullary SUVmax (r=0.620) were positively correlated with the body temperature. After normalized by SUV of blood pool or liver, the differences of SUV in brain, liver or blood pool between two groups were still statistically significant (P<0.05).

Conclusions: Our results indicate that hyperthermia causes significantly alteration in the biodistributions of ¹⁸F-FDG in various organs. Changes of SUVs may affect the detection of lesions as well as the evaluation of therapeutic response in tumor. We suggest that patients with high body temperature are not suitable for PET/CT. The SUV normalized by blood pool or liver SUV does not mitigate the impact of body temperature in the course of evaluating the lesion or the therapeutic response with ¹⁸F-FDG PET/CT in febrile patients.Keyword: ¹⁸F-FDG biodistribution; PET/CT; standardized uptake value; fever; therapeutic response evaluation