The Research For Physiological Distribution Of The Spinal Cord In ¹⁸F-FDG PET/CT

Research background and purpose:Positron emission tomography/computer tomography（PET/CT）is an imaging device that combines PET（functional metabolic imaging）and CT（anatomical imaging）.And ¹⁸F-FDG is a broad-spectrum PET/CT imaging agent.In ¹⁸F-FDG PET/CT imaging,the physiological uptake of various normal organs or tissues can often be easily confused with pathological uptake.To accurately interpret PET/CT images,it is important to be familiar with the normal sites of ¹⁸F-FDG distribution throughout the body,the intensity of uptake and the factors that may influence it.In the case of the spinal cord,it is often necessary to use PET/CT to diagnose spinal metastases in patients with tumours,which need to be distinguished from physiological uptake.Given the paucity of studies on the distribution of physiological tracers in the spinal cord and their identification with spinal metastases,this study collected clinical data and PET/CT image data from 83 healthy adult subjects and 10 patients with spinal metastases for cross-sectional analysis and comparison,with the following main aims:Part 1:To assess the pattern and characteristics of physiological ¹⁸F-FDG distribution in the spinal cord in ¹⁸F-FDG PET/CT and to explore the mechanisms associated with physiological uptake in the spinal cord,with a view to improving the accuracy of PET/CT images and providing an evidence-based basis for clinical diagnosis and differential diagnosis.Part 2:To evaluate the feasibility and value of ¹⁸F-FDG PET/CT for the differential diagnosis of physiological spinal cord uptake and spinal cord metastases,and to explore the best predictive value of each metabolic parameter in PET/CT for the differentiation of increased spinal cord uptake foci from metastases and physiological ones.Materials and Methods:Part 1:A total of 83 subjects who underwent whole-body ¹⁸F-FDG PET/CT physical examinations at our nuclear medicine department PET/CT centre between April 13,2018 and June 30,2022 were collected according to the inclusion and exclusion criteria.The maximum standardized uptake value（SUVmax）of the spinal cord at each vertebral level was collected from the PET-CT fusion images,and the mean standardized uptake value（SUVmean）of the liver was used as the background.Record the target-to-background ratio（TBR）（defined as spinal SUVmax/liver SUVmean at each vertebral level）,and the aim is to avoid the effect of substrate on spinal cord uptake,so the spinal cord SUVmax was normalised using the liver as the substrate..Pearson’s test and Spearman’s rank correlation analysis were used to determine the correlation between spinal cord full-segment TBR,TBR at each vertebral level and spinal cord cross-sectional area,age,weight,fasting blood glucose,dose administered per kilogram of body weight,time of administration,temperature,and the correlation between maximum TBR in the cervical and thoracic segments;two independent sample t-tests/corrected t-tests were used to compare male and female spinal cord full-segment TBR,differences in TBR at each vertebral level.A difference of P<0.05 was considered statistically significant.Part 2:A total of 10 patients with spinal cord metastases who underwent whole-body ¹⁸F-FDG PET/CT at our nuclear medicine department between July 24,2015 and November 2,2022 were collected,and 10 spinal cord physiological uptake controls were also included.The SUVmax,SUVmean,Peak standardized uptake value（SUVpeak）,Maximum standardized uptake value corrected for lean body mass（SULmax）,Mean standardized uptake value corrected for lean body mass（SULmean）,Peak standardized uptake value corrected for lean body mass（SULpeak）,metabolic tumor volume（MTV）,total lesion glycolysis（TLG）were collected from focal foci of increased spinal uptake in both groups and their target to book ratio TBR（SUVmax/Liver SUVmean）was recorded.P<0.05 was considered a statistically significant difference.For focal foci of increased uptake in both groups,the area under the curve（AUC）was calculated using the subject operator characteristic curve（ROC）and the Youden’s index was used to determine the metastases and physiological The optimal cut-off values for SUVmax,SUVmean,SUVpeak,SULmax,SULmean,SULpeak,MTV,TLG and TBR were determined using the Youden’s index.Results:Part 1:1.83 subjects were able to observe two focal foci of increased uptake at the level of the lower cervical and lower lumbar spine on PET/CT images.The mean TBR in healthy subjects showed a roughly decreasing trend along the spinal cord from the cervical to the lumbar spine and reached two peaks in the lower cervical spine（at the C4 vertebrae level）and the lower thoracic spine（at the T11-12 vertebrae level）,respectively.The mean TBR of the cervical spinal cord ranged from 1.133±0.233 at the C2 vertebral level to 1.181±0.255 at the C4 vertebral level,and then decreased to0.645±0.146 at the T7 vertebral level.however,the relative increase at the C4 vertebral level was not significant.the overall TBR at the T8-T10 vertebral level did not change significantly.There was a significant relative increase in TBR at the T11 and T12vertebral levels,with TBR values reaching 0.927±0.214 and 1.134±0.288 respectively.2.There was an extremely strong positive correlation between spinal cord cross-sectional area and TBR,and an extremely strong positive correlation between maximum TBR in the cervical and thoracic segments（P<0.05）.3.Age did not correlate with TBR of the whole spinal cord and there was no correlation between age and TBR at any vertebral level（P>0.05）.4.Body weight had a mild negative correlation with spinal cord total TBR（r=-0.223,P=0.042）.There was also a mild negative correlation between body weight and TBR at several vertebral levels:C3（r=-0.257,P=0.019）、C5（r=-0.216,P=0.049）、C6（r=-0.303,P=0.005）、C7（r=-0.271,P=0.013）、T1（r=-0.217,P=0.049）、T4（r=-0.223,P=0.043）、T5（r=-0.258,P=0.018）、T7（r=-0.254,P=0.021）、T8（r=-0.229,P=0.037）;all other vertebral levels were not significantly different（P>0.05）.5.Fasting blood glucose had a mild negative correlation with spinal cord full-segment TBR（r_s=-0.284,P=0.009）.and there was a mild negative correlation between fasting blood glucose and TBR at multiple vertebral levels:C2（r_s=-0.284,P=0.009）、C3（r_s=-0.227,P=0.039）、C4（r_s=-0.286,P=0.009）、C5（r_s=-0.239,P=0.03）、C6（r_s=-0.288,P=0.008）、C7（r_s=-0.255,P=0.02）、T3（r_s=-0.277,P=0.011）、T4（r_s=-0.35,P=0.001）、T5（r_s=-0.307,P=0.005）、T7（r_s=-0.293,P=0.007）、T12（r_s=-0.318,P=0.003）;all other vertebral levels were not significantly different（P>0.05）.6.There was no correlation between dose per kg body weight administered and TBR for the whole spinal cord and no correlation between dose per kg body weight administered and TBR at any vertebral level（P>0.05）.7.There was a moderate positive correlation between time of administration and spinal cord total TBR（r=0.495,P<0.001）and a mild/moderate positive correlation between time of administration and TBR at all vertebral levels.8.There was no correlation between temperature and TBR in the whole spinal cord.However,there was a mild positive correlation between temperature and TBR at the T3 vertebral level（r_s=0.221,P=0.044）.9.At a total of 19 vertebral levels from C2-L1,TBR was higher in females than in males at all 17 vertebral levels,except at T10 and L1 vertebral levels.There was a significant difference in spinal cord whole segment uptake between males and females and a significant difference in TBR at multiple vertebral levels（P<0.05）.Part 2:1.In the 10 patients in the spinal cord metastasis group,seven presented with single focal foci of increased uptake in the spinal cord and three presented with multiple focal foci of increased uptake in the spinal cord.A total of 25 focal foci of increased uptake were observed in the metastatic group.The mean SUVmax value was6.488±4.068,the mean SUVmean value was 3.913±2.304,the mean SUVpeak value was 4.061±2.643,the mean SULmax value was 5.466±3.382,the mean SULmean value was 3.298±1.918,the mean SULpeak value was 3.5±2.068,MTV mean value was3.374±4.384,TLG mean value was 12.963±23.192 and TBR mean value was3.275±2.166.2.In the 10 subjects in the physiological uptake group,two focal foci of increased uptake were observed in all subjects at the cervical and lumbosacral expansions of the spinal cord.A total of 20 focal foci of increased uptake were observed in the physiological uptake group.In the physiological uptake group,the mean SUVmax value was 2.641±0.377,the mean SUVmean value was 1.802±0.325,the mean SUVpeak value was 2.126±0.299,the mean SULmax value was 2.027±0.25,the mean SULmean value was 1.384±0.21,the mean SULpeak value was 1.644±0.188 and the mean MTV value was 1.644±0.188.0.188,MTV mean value was 1.456±0.676,TLG mean value was 2.661±1.321 and TBR mean value was 1.223±0.235.3.The differences between SUVmax,SUVmean,SUVpeak,SULmax,SULmean,SULpeak,TLG,and TBR were statistically significant（P<0.001）for spinal metastases,and physiological uptake,with z-values of-4.614,-3.86,-3.883,-4.715,-4.203,-4.431,-3.974,-5.231.while the difference in MTV between spinal transfer,physiological uptake was not statistically significant（P>0.05）.SUVmax,SUVmean,SUVpeak,SULmax,SULmean,SULpeak,TLG and TBR were all greater than physiological uptake for spinal cord transfer.4.The ROC curves showed that the best cut-off values for spinal transfer and physiological uptake were SUVmax=3.316,AUC=0.904,P<0.001;SUVmean=2.372,AUC=0.838,P<0.001;SUVpeak=2.551,AUC=0.84,P<0.001;SULmax=2.537,AUC=0.916,P<0.001;SULmean=1.743,AUC=0.868,P<0.001;SULpeak=1.969,AUC=0.888,P<0.001;TLG=3.809,AUC=0.848,P<0.001;TBR=1.609,AUC=0.958,P<0.001.Conclusions:1.Spinal physiological FDG uptake in healthy adult subjects showed a roughly decreasing trend from the cervical to the lumbar spine,with two peaks at the level of the lower cervical spine（C4）and the lower thoracic spine（T11-T12）,respectively,which may be closely related to cervical and lumbosacral dilation.2.Body weight,fasting blood glucose,dose per kilogram of body weight,time of administration,and gender were all correlated with physiological spinal cord uptake,whereas age and temperature were not significantly correlated with physiological spinal cord uptake.3.PET/CT semi-quantitative metabolic parameters（SUVmax,SUVmean,SUVpeak,SULmax,SULmean,SULpeak,MTV,TLG,TBR）were all statistically significant in identifying spinal cord metastases and physiological uptake,with the best cut-off values being:SUVmax=3.316,SUVmean=2.372,SUVpeak=2.551,SULmax=2.537,SULmean=1.743,SULpeak=1.969,TLG=3.809,TBR=1.609.