Abstract
Objectives
18Fluorine-fluorodeoxyglucose (18F-FDG) uptake in the spinal cord is not unusual and may mimic metastatic disease. The physiological characterization and variability of spinal cord 18F-FDG metabolism provide valuable information, especially in patients with suspected malignancies. We aimed to investigate the physiological 18F-FDG uptake pattern within the spinal cord and its associations in a normal population.
Methods
We retrospectively analyzed 18F-FDG positron emission tomography/computed tomography images of 140 adult patients who were confirmed to be disease-free over a one-year follow-up period. The maximal and mean standard uptake values (SUVmax, SUVmean) were measured at each mid-vertebral level from C1 to L5, and normalized to liver and blood pool uptake. Correlations between 18F-FDG uptake and patient demographics, clinical parameters, and environmental temperature were evaluated.
Results
18F-FDG uptake demonstrated a decreasing trend from the cervical to lumbar vertebrae, with a notable increase at the lower thoracic levels (T11-T12). There was a significant negative correlation between 18F-FDG uptake and age (p<0.001), fasting glucose level (p=0.016), and diabetic status (p=0.003). No significant association was found between spinal cord <sup>18</sup>F-FDG uptake and gender, weight, height, body mass index, 18F-FDG dose, or environmental temperature.
Conclusion
Normal distribution of 18F-FDG in the spinal cord of disease-free individuals decreases from cervical to lumbar levels, although it notably increases at the lower thoracic and mid-lower cervical levels. Uptake significantly decreases with age, with a higher fasting blood glucose level, and in diabetic patients.
Introduction
18Fluorine-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) imaging is increasingly being used in the initial diagnosis, staging, restaging, and monitoring therapy response in oncological patients. The physiological distribution of 18F-FDG has been welldefined; it has high uptake in the heart, liver and brain, eliminates in urine, and concentrates in the gastrointestinal tract with varying degrees. 18F-FDG uptake in the spinal cord is not unusual and may be either physiological or indicative of underlying diseases such as metastatic involvement, vascular, infectious, or inflammatory diseases (1). Spinal metastases constitute approximately 8.5% of all the central nervous system metastases, with lung and breast cancer being the most common primaries (2). Normal variants of physiological uptake in the spinal cord can cause misdiagnosis and unnecessary treatments in oncologic patients. Hence, it is crucial to understand the physiologic distribution of 18F-FDG in the spinal cord for a correct interpretation of PET imaging.
Several studies examined the physiological distribution of 18F-FDG within the spinal cord of adult cancer patients with non-central nervous system malignancies (1,3,4,5,6,7,8), while two studies evaluated the physiological metabolism of the spinal cord in oncologic pediatric patients (9,10). However, these studies have confounding factors such as chemotherapy, radiation therapy, or surgery, that might affect 18F-FDG uptake in the spinal cord due to radiation myelopathy (11,12). To date, only two studies with small sample sizes (n=16, n=30) reported physiologic uptake of 18F-FDG within the spinal cord in disease-free patients (13,14). The aim of this study is to evaluate the physiological 18F-FDG distribution within the spinal cord in disease-free subjects, and assess the influence of gender, age, body weight, body mass index (BMI), diabetes and environmental temperature on spinal cord 18F-FDG uptake in a larger cohort population.
Materials and Methods
Study Design and Patient Selection
We retrospectively evaluated 18F-FDG PET/CT images of adult patients who were referred to our clinic for evaluating suspected malignancy, or characterization of solitary pulmonary nodules over a 24-month period from June 2021 to June 2023. Only 140 patients who were proven to be disease-free during a one-year follow-up were included in the study. Patients with abnormal 18F-FDG PET/CT findings suggestive of malignant disease, a history of former malignancies, chemotherapy, inflammatory or degenerative vertebral diseases affecting spinal cord, and previous spinal operations were excluded from participating in the study. Additionally, patients whose fasting blood glucose levels were higher than 126 mg/dL were excluded, in accordance with the European Association of Nuclear Medicine procedural guidelines for research studies (15). Medical parameters including gender, age, weight, height, BMI, fasting glucose level, diabetes status, and 18F-FDG dose were recorded. Daily local average temperature values were obtained from the national meteorology archives. This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Approval was obtained from the Marmara University Faculty of Medicine Clinical Research Ethics Committee (number: 09.2021.1437, date: 03.12.2021) and informed consent was obtained from each participant.
18F-FDG PET/CT protocol
18F-FDG PET/CT images were acquired after a 6-hour fast and approximately 1 hour after intravenous injection of 18F-FDG using a dedicated combined scanner (GE Discovery ST; GE Healthcare, Milwaukee, WI). All patients had a blood sugar level of less than or equal to 126 mg/dL before 18F-FDG injection. First, a multi-slice CT scan was performed with a 16-slice multidetector scanner (parameters: 80 mA; 140kV; table speed: 27 mm/rotation; and slice thickness: 3 mm) from the top of the head through the feet in the supine position in a shallow breathing patient. A routine whole-body PET scan was conducted in 3D mode, with an acquisition time of 3 minutes per bed position, covering the same area as the CT scan. PET data were reconstructed using an iterative processing algorithm, and the acquisition data transferred to a workstation (Advantage Windows Server 4.5; GE Healthcare) for manual segmentation and interpretation.
Image Analysis
All 18F-FDG PET/CT images were evaluated by two experienced nuclear medicine physicians.
The maximal and mean standard uptake values (SUVmax; SUVmean) of the spinal cord at each mid-vertebral level from C1 to L5 were recorded with a standard region of interest (ROI) size. The ROIs for each spinal cord measurement were manually drawn while avoiding the margins of vertebrae (Figure 1). For internal normalization, a reference ROI was placed over the liver, on the 6th hepatic segment, and over the blood pool, particularly on the right atrium, to calculate the normalized SUVmax (nSUVmax) and the normalized SUVmean (nSUVmean) values.
Statistical Analysis
Descriptive statistics were used to present the characteristics of the study population. Continuous data were reported as means ± standard deviations or median, (range), while categorical variables were expressed as frequencies (percentage). A preliminary Kolmogorov-Smirnov test was used to assess the normality of variables. Means were compared with Student’s t-test. The correlations of SUV with age, weight, 18F-FDG dose, and temperature were tested with Spearman’s correlation analysis. Pearson’s correlation analysis was used to determine the relationship between glucose level, height, BMI, and the 18F-FDG uptake of each vertebral level. The pattern of physiological spinal cord, 18F-FDG, distribution was determined by drawing a graph based on the mean nSUVmax/nSUVmean of each vertebral level. Data were analyzed using IBM SPSS® software version 29.0. Results with two-sided p<0.05 were considered statistically significant.
Results
Patient Characteristics
One hundred forty patients (72 men, 68 women) who met the inclusion criteria and were enrolled in the study. The mean age of the patients was 55±16 (men 54.9±16, women 55.2±16.2) years, while the median age was 53.5. Population characteristics are summarized in Table 1.
SUV Measurements
The highest SUVmax value was noted at the C1 level (range 1.3-3.1, mean 2.1), while the lowest SUVmax value was observed at the L5 level (range 0.5-2.4, mean 1.1). Physiologic 18F-FDG uptake showed a decreasing pattern in the spinal cord from cervical to lumbar vertebrae with a significant increase at the lower thoracic (T11-T12) levels and a relatively insignificant increase at mid-cervical (C4) level. It is notable that the SUVmax and SUVmean values, as well as nSUVmax and nSUVmean values, were highly correlated (p<0.000). The mean SUVmax of the spinal cord at the C1 and T12 levellevels was 2.1 and 1.8, respectively. Normal distribution of maximal spinal cord SUV measurements is presented in Figure 2.
Factors affecting 18F-FDG uptake in the spinal cord
There was a statistically significant association between spinal cord 18F-FDG uptake intensity and the presence or absence of diabetes (p=0.003). Also, spinal cord 18F-FDG uptake had a negative correlation with age and fasting glucose level (r=-0.37, p<0.001 and r=-0.20, p=0.016, respectively, Figure 3). However, no significant association was found between spinal cord 18F-FDG uptake intensity and gender (p=0.27), patient weight (p=0.29), height (p=0.51), BMI (p=0.44), 18F-FDG dose (p=0.70), or environmental temperature (p=0.27). Figures 4 and 5 demonstrate that there is a decrease in 18F-FDG uptake in the spinal cord in patients with diabetes and with increasing age, respectively.
Discussion
Our results confirm that spinal cord 18F-FDG uptake significantly correlates with patients’ age, fasting blood glucose level and diabetic status. No significant association was found between SUV and gender, patient weight, height, BMI, 18F-FDG dose, or environmental temperature. This study also shows the physiological distribution of 18F-FDG uptake within the spinal cord, which is more evident in the cervical and lower thoracic levels. The main difference in our study lies in the study population, which is therapy-naive and proven to be disease-free with the largest sample size reported in the literature.
Several studies reported spinal cord 18F-FDG distribution with a variety of investigation methods. Only two studies calculated SUVmax of all spinal segments and reported a decreasing pattern from cervical to lumbar levels with an increase at the lower thoracic and mid-lower cervical levels, which complies with our results (1,14). This pattern is attributed by the authors to cervical and lumbar enlargement and the increased amount of gray matter in children and adults (1,6,9). Another aspect of our study was to evaluate the consistency of SUV calculated within the spinal cord and those normalized to liver or blood pool. The physiological distribution of SUVmax and SUVmean in the spinal cord correlated well with the distribution of nSUVmax and nSUVmean.
Most of the studies reported in the literature were carried out in patients with a history of cancer and radiation therapy. However, white matter necrosis, demyelination, and malacia are typical features of radiation damage to the spinal cord, which may result in decreased 18F-FDG uptake (16). In addition, radiation myelopathy is associated with increased 18F-FDG uptake in the irradiated spinal cord (11,12). Although some studies excluded patients who had previously received radiation therapy, the studies did not consider the impact of chemotherapy, which could also interrupt the blood-spinal cord barrier, resulting in edema, demyelination, and finally necrosis and atrophy of the spinal cord (17).
Two pediatric studies reported a significant increase in SUV within the spinal cord with increasing age (9,10). In adult patients, most studies did not find any correlation between spinal cord 18F-FDG uptake and age (1,3,5,14,18). Contrary to these studies, Kamoto et al. (19) and Tan et al. (20) reported a negative association between metabolic activity of the spinal cord and age in oncologic patients. Our study is the first to determine a negative correlation between age and spinal cord metabolic activity in a healthy adult population. Although neural plasticity and volumetric growth of the spinal cord during childhood development could be responsible for the positive correlation between age and spinal cord activity, age-related decreases in SUV in adult patients may be a reflection of age-associated structural atrophy, reduced nerve conduction velocity, and reduced motor activity (21).
Greenspan et al. (7) reported a statistically significant association between distal spinal cord uptake and lower blood glucose levels; however, there was no significant association between patients’ diabetic status and any other variable. In 2022, Tan et al. (20) reported that intense 18F-FDG uptake in the distal spinal cord was more common in patients without diabetes and with lower blood glucose levels. Contrarily, the studies conducted by Nakamoto et al. (18) and Patel et al. (5) did not find a correlation between the plasma glucose level and the cervical spinal cord 18F-FDG uptake. In our study, mean spinal cord SUV and nSUV values were lower in diabetic patients, who also exhibited higher glucose levels. Previous studies reported lower cerebral 18F-FDG uptake in patients with diabetes (22). Although the pathogenesis is not fully understood, insulin-induced upregulation of hepatic glucokinase results in hepatocellular glucose uptake and increased glucose flux to the central nervous system. While hyperglycemia is reported to have a greater impact on 18F-FDG uptake in the central nervous system compared to the liver (23), the reduced SUV and nSUV in our study group may be explained by poorly controlled hyperglycemia in diabetes, leading to competitive inhibition of 18F-FDG uptake in normal tissues.
In this study, no association was found between patients’ weight, height, or BMI and spinal cord metabolism. A few studies reported a positive correlation between body weight and spinal cord 18F-FDG uptake (6,10). Since 18F-FDG distribution is very low in adipose tissue, higher SUV is expected in overweight patients. This vague result could be mitigated by applying SUV correction for lean body mass or body surface area.
With regard to seasonal variation, Amin et al. (3) reported increased cord 18F-FDG uptake in winter, which is based on the hypothesis that activation of the sympathetico-adrenal system leads glucose flux to the central nervous system. However, similar to the results reported by other investigators, we did not find a relation between environmental temperature and spinal cord 18F-FDG uptake (4,10,14).
An association between gender and cord uptake has been investigated by many authors. Greenspan et al. (7) and Taralli et al. (10) reported a positive correlation with female sex. This can be explained by the higher metabolism of the lumbosacral tract secondary to women’s reproductive system innervation. However, sex difference was not reported in most other studies, which are similar to our results (3,5,14,18,20).
The limitations of this study include its retrospective design. Also, the time interval between 18F-FDG injection and imaging was not taken into account in correlation analyses. Nevertheless, the strong points of our study are the SUV measurement method, which includes each mid-vertebral spinal level, and the inclusion of a specific disease-free patient group.
Conclusion
In normal subjects, physiological 18F-FDG uptake in the spinal cord decreases from cervical to lumbar levels, although there is an increase noted at the lower thoracic and mid-lower cervical regions. The present study provides evidence that spinal cord 18F-FDG uptake significantly decreases with increasing age, blood glucose level, and diabetes status in patients. Even though our study included the largest number of normal adult participants to date, the conflicting results with other studies reported in the literature warrant further research with larger sample sizes to obtain more conclusive results.
