May 2008, VOLUME 190
NUMBER 5

Recommend & Share

May 2008, Volume 190, Number 5

Nuclear Medicine

Original Research

Method for Decreasing Uptake of 18F-FDG by Hypermetabolic Brown Adipose Tissue on PET

+ Affiliation:
1Both authors: Department of Radiology, Division of Nuclear Medicine, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215.

Citation: American Journal of Roentgenology. 2008;190: 1406-1409. 10.2214/AJR.07.3205

ABSTRACT
Next section

OBJECTIVE. The purpose of this study was to determine whether use of a high-fat, very-low-carbohydrate protocol for preparing patients for PET decreases the frequency of 18F-FDG uptake by hypermetabolic brown adipose tissue (BAT) on PET scans.

MATERIALS AND METHODS. In this HIPAA-compliant retrospective study, 741 FDG PET/CT scans obtained during the winter months (October 1–April 30) for patients who prepared with a high-fat, very-low-carbohydrate, protein-permitted protocol were compared with 1,229 FDG PET scans obtained during the winter months for patients who prepared by fasting. FDG uptake on PET scans co-localized with regions of fat identified on the CT scans was assumed to represent hypermetabolic BAT. The categoric variables frequency of occurrence of hypermetabolic BAT (present or not) and the sex ratios of the groups before and after the change in preparation were compared by use of a chi-square test. The continuous variables of age and blood glucose level were compared by use of a two-tailed Student's t test.

RESULTS. In this intention-to-treat analysis, there was no difference between the fasting (n = 1,229) and the high-fat, very-low-carbohydrate, protein-permitted diet (n = 741) groups in terms of age and sex. Patients who prepared with the high-fat diet had a significantly lower frequency of hypermetabolic BAT uptake on FDG PET scans during the winter months (p<0.0002) and had lower blood glucose levels (p≪0.001).

CONCLUSION. In this intention-to-treat analysis, use of a high-fat preparation protocol significantly lowered the frequency of uptake of FDG by hypermetabolic BAT on FDG PET studies. Use of this protocol has the potential to decrease the rate of false-positive findings on oncologic FDG PET scans.

Keywords: brown adipose tissue, brown fat, FDG PET, high-fat low-carbohydrate diet, patient preparation

Introduction
Previous sectionNext section

Brown adipose tissue (BAT) is histologically and physiologically distinct from white adipose tissue and is thought to be involved in thermogenesis rather than fat storage [1, 2]. On 18F-FDG PET scans, hypermetabolic BAT is often but not exclusively recognized as symmetric uptake in the cervical, mediastinal, and paraspinal areas [3, 4]. This FDG uptake by hypermetabolic BAT can be confused with nodal uptake, leading to false-positive results [5]. In one series [6], 45% of patients undergoing PET for breast cancer had FDG uptake consistent with hypermetabolic BAT in terms of distribution. FDG uptake by hypermetabolic BAT not only can lead to false-positive findings but also can decrease the sensitivity of tumor uptake by decreasing the pool of FDG available [7]. One net effect is that uptake by hypermetabolic BAT can be misidentified as a sign of metastatic disease. The other effect is that because it is difficult to exclude uptake as due to hypermetabolic BAT rather than tumor, the finding of uptake can lead to unnecessary biopsy or repeated scanning with additional radiation exposure.

Several methods to have been tried to reduce FDG uptake by hypermetabolic BAT. None has been reliable, and most have involved pharmacologic intervention, which is especially undesirable in the pediatric population. Warming has been used, as has administration of benzodiazepines, propranolol, and even fentanyl [811]. Nicotine has been found to increase uptake of hypermetabolic BAT [12], prompting the suggestion that nicotine be avoided before FDG injection. A reliable nonpharmacologic method clearly is desirable.

Hypermetabolic BAT contains considerable cytoplasm, multiple small lipid droplets, round centrally located nuclei, and abundant mitochondria. In contrast, white adipose cells contain a scant ring of cytoplasm around a single large lipid droplet, an eccentric flattened nucleus, and essentially no mitochondria [13]. Hypermetabolic BAT requires a high mitochondrial concentration for thermogenesis [1, 2]. We [14] have investigated another mitochondria-rich tissue, the myocardium. In the myocardium, the high mitochondrial concentration produces adenosine tri phosphate for muscle fiber contraction. We found that consumption of a high-fat, very-low-carbohydrate, protein-permitted diet 3–5 hours before FDG injection minimized myocardial FDG uptake [14]. Results of a study of the Randle cycle (fatty acid–glucose cycle) have shown that fatty acid loading suppresses glucose metabolism [15]. We reasoned that the timing allowed fatty acids and triglycerides to peak after the postprandial insulin peak, inhibiting glycolysis through the Randle effect and providing the preferred myocardial fuel [15]. We also reasoned that this effect might be realized in another mitochondria-rich tissue, hypermetabolic BAT. In this tissue, energy is not used for muscle contraction but is dissipated as heat. We undertook a retrospective study to investigate whether the change in preparation from an overnight fast to a high-fat, very-low-carbohydrate, protein-permitted diet would decrease the frequency with which hypermetabolic BAT FDG uptake is observed.

Materials and Methods
Previous sectionNext section

This HIPAA-compliant retrospective study was approved by the institutional review board at our institution. All patients underwent FDG PET for oncologic indications. FDG PET/CT scans were reviewed to determine the frequency of the presence of hypermetabolic BAT before (fasting) and after (high-fat, very-low-carbohydrate, protein-permitted diet) a change was made to the clinical protocol to minimize glucose and insulin levels. Areas of FDG uptake on PET scans that co-localized with regions of fat identified on the CT scans were assumed to represent hypermetabolic BAT.

Winter was defined as October 1 through April 30. For winters 2002–2003 and 2003–2004, 1,229 FDG PET scans were analyzed for patients who prepared for imaging with an overnight fast. For winter 2005–2006, 741 FDG PET scans were analyzed for patients who prepared by eating a high-fat, very-low-carbohydrate, protein-permitted diet on the night before and on the morning of imaging. Recommended foods included chicken, turkey, fish, meat (e.g., steak, ham), meat-only sausages, fried eggs, bacon, scrambled eggs prepared without milk, omelet prepared without milk or vegetables, fried eggs and sausages, and fried eggs and bacon. Patients were directed to avoid sugar and other carbohydrates, milk, cheese, bread, bagels, cereal, cookies, toast, pasta, crackers, muffins, peanut butter, nuts, fruit juice, potatoes, candy, fruit, rice, chewing gum, mints, cough drops, vegetables, beans, and alcohol [14]. There was no change in other instructions to the patients, including avoiding cold exposure.

Data on sex, age, and blood glucose level were collected. Frequency of presence of hypermetabolic BAT and sex ratio of the groups, the categoric variables, before and after the change in patient preparation were compared by use of a chi-square test. Age and blood glucose level, the continuous variables, were compared by use of Student's t test. One investigator, who was not blinded, used a visual analog grading system to evaluate hyper-metabolic BAT. Absence of hyper metabolic BAT was scored 0, and presence was scored 1–6 on the basis of the extent and amount of FDG uptake (Fig. 1A, 1B, 1C, 1D). CT was used to confirm the density and expected anatomic location of hypermetabolic BAT.

After IV injection of 740 MBq (20 mCi) of FDG, the patients were kept in a semidarkened quiet room for 45–60 minutes. Plasma glucose mea surements before injection were less than 200 mg/dL in all patients. Imaging was performed with a 4-MDCT PET/CT scanner (Discovery/LightSpeed, GE Health care). A CT scout image (30 mA, 120 kVp) was obtained. Scanning then was performed from the base of the skull to midthigh with helical CT at 0.8 s/rotation at 100 mA and 149 kVp with a section thickness of 5 mm and a 4.25-mm interval. Thin oral barium but no IV contrast material was used in all studies. Patients were instructed to breathe normally during acquisition. PET images were obtained at 5 minutes per bed position. PET images were iteratively reconstructed with CT-based attenuation correction.

Results
Previous sectionNext section

Age, sex, blood glucose level, and frequency of the presence of FDG uptake by hyper-metabolic BAT are shown in Table 1. All patients, even those who did not adhere to the instructions, were included in this intention-to-treat analysis.

TABLE 1: Comparison of Age, Sex, Blood Glucose Level, and Frequency of Presence of 18F-FDG Uptake by Hypermetabolic Brown Adipose Tissue Before and After Change in Patient Preparation

Figure 1A, 1B, 1C, 1D shows the grading system used to assess uptake by hypermetabolic BAT. A grading system of 0–6 was used. The absence of detectable uptake by hypermetabolic BAT was graded 0. Faint uptake was graded 1, and widespread uptake in the neck, supraclavicular areas, and paraspinal areas was designated 6. Figure 2A shows uptake on a PET scan; Figure 2B, a CT scan at the same level as Figure 2A; and Figure 2C, a fused image depicting FDG uptake in the areas of fat seen on the CT scan. The winter serial scans of one patient before and after changes in the preparation protocol are shown in Figure 3A, 3B, 3C.

The fasting protocol (n = 1,229) and high-fat, very-low-carbohydrate, protein-permitted diet (n = 741) groups did not differ in terms of age and sex. There was a statistically significant difference between the groups in terms of blood glucose level and frequency of FDG uptake by hypermetabolic BAT, which decreased from 6.3 to 2.8 patients per 100 on an intention-to-treat basis. Patients prepared with the high-fat diet had a significant decrease in frequency with which hypermetabolic BAT uptake occurred on FDG PET scans (p<0.0002), and these patients had lower blood glucose levels (p≪0.001).

Discussion
Previous sectionNext section

The significant decrease in frequency of the presence of uptake of FDG by hypermetabolic BAT on winter-month PET scans of patients who prepared with the high-fat, very-low-carbohydrate, protein-permitted diet parallels the decrease in FDG uptake by another mitochondria-rich tissue, the myocardium, with use of the same high-fat protocol [14]. This decrease is probably the result of two factors. First, hypermetabolic BAT has a high capacity for glucose utilization and is a major site for lipid metabolism. Fatty acids are the main fuel for maintaining the thermogenic capability of hypermetabolic BAT, in which hormone-sensitive lipase is used to obtain fatty acids from triglycerides, and lipoprotein lipase is used to capture fatty acids from lipoproteins [16, 17].

figure
View larger version (99K)
Fig. 1A PET/CT scans show scores of uptake of 18F-FDG by brown adipose tissue. In grading system, score of 0 represented no uptake; 6, very intense uptake. Numbers on images indicate scores. 40-year-old woman with history of large cell lymphoma to thyroid gland; score is 1.

figure
View larger version (121K)
Fig. 1B PET/CT scans show scores of uptake of 18F-FDG by brown adipose tissue. In grading system, score of 0 represented no uptake; 6, very intense uptake. Numbers on images indicate scores. 41-year-old woman with non-Hodgkin's lymphoma who was seen for restaging; score is 3.

figure
View larger version (109K)
Fig. 1C PET/CT scans show scores of uptake of 18F-FDG by brown adipose tissue. In grading system, score of 0 represented no uptake; 6, very intense uptake. Numbers on images indicate scores. 76-year-old woman undergoing staging for lymphoma; score is 4.

figure
View larger version (118K)
Fig. 1D PET/CT scans show scores of uptake of 18F-FDG by brown adipose tissue. In grading system, score of 0 represented no uptake; 6, very intense uptake. Numbers on images indicate scores. 53-year-old woman with history of breast cancer who presented with right lung nodule; score is 6.

figure
View larger version (44K)
Fig. 2A 34-year-old woman who underwent restaging of lymphoma. PET scan shows areas of brown adipose tissue.

figure
View larger version (67K)
Fig. 2B 34-year-old woman who underwent restaging of lymphoma. CT scan confirms areas of brown adipose tissue by fat density and location.

figure
View larger version (65K)
Fig. 2C 34-year-old woman who underwent restaging of lymphoma. Fused PET/CT scan shows 18F-FDG uptake corresponding to fat density in B.

figure
View larger version (75K)
Fig. 3A 39-year-old woman with Hodgkin's disease who presented for restaging. Serial 18F-FDG PET/CT scans for which patient prepared by fasting in February 2004 (A) and December 2004 (B) and with high-fat, very-low-carbohydrate, protein-permitted diet in November 2005 (C) show suppression of FDG uptake by hypermetabolic brown adipose tissue.

figure
View larger version (78K)
Fig. 3B 39-year-old woman with Hodgkin's disease who presented for restaging. Serial 18F-FDG PET/CT scans for which patient prepared by fasting in February 2004 (A) and December 2004 (B) and with high-fat, very-low-carbohydrate, protein-permitted diet in November 2005 (C) show suppression of FDG uptake by hypermetabolic brown adipose tissue.

figure
View larger version (78K)
Fig. 3C 39-year-old woman with Hodgkin's disease who presented for restaging. Serial 18F-FDG PET/CT scans for which patient prepared by fasting in February 2004 (A) and December 2004 (B) and with high-fat, very-low-carbohydrate, protein-permitted diet in November 2005 (C) show suppression of FDG uptake by hypermetabolic brown adipose tissue.

Second, in the Randle cycle (fatty acid–glucose cycle) fatty acid loading inhibits glucose metabolism [15]. Mitochondrial B-oxidation of fatty acids supplies the proton gradient, which is uncoupled by the uncoupling protein UCP-1 to provide energy that is dissipated as heat in the process called thermogenesis [1820]. The Randle effect is multifactorial but in part involves decreased pyruvate dehydrogenase activity, which inhibits glucose oxidation, and inhibition of phosphofructokinase through an increase in citrate so that glucose-6-phosphate levels increase. The Randle effect also involves inhibit ion of hexokinase activity, leading to de creased glucose utilization. This effect may explain the lower blood glucose levels in the group prepared for PET/CT with a high-fat, very-low-carbohydrate, protein-permitted diet. Further work is under way to determine the exact physiologic mechanisms involved.

After searching the literature, we believe that our results are the first showing that a high-fat diet before FDG PET/CT decreases the frequency of detection of FDG uptake by hypermetabolic BAT on PET scans. This method of patient preparation may be a means of decreasing false-positive findings on oncologic FDG PET scans. Given the volume of FDG PET performed at oncologic centers and the baseline occurrence of hypermetabolic BAT, which varies among centers, use of the high-fat preparation may prevent repeated scanning and the need to declare limitations to sensitivity in 12 (200 scans per month with a 6% incidence of hypermetabolic BAT) to 450 (1,000 scans per month with a 45% incidence of hypermetabolic BAT) imaging procedures per month at a cost savings of $12,000–$450,000 per month (technical, radiopharmaceutical, and professional fees).

There were limitations to this study. First, the exact mechanism by which a high-fat diet 3–6 hours before FDG injection decreases FDG uptake by hypermetabolic BAT is not unambiguously defined. Second, we used a timed high-fat meal that was difficult to standardize and involved patient adherence. We are investigating a timed, standardized high-fat preparation that in initial evaluation shows stronger suppression and more reliability in decreasing both myocardial and hypermetabolic BAT uptake of FDG. Nonetheless, in this intention-to-treat study, hypermetabolic BAT suppression was marked. We would therefore expect that if there were increased adherence to the high-fat, very-low-carbohydrate, protein-permitted diet through use of a standardized preparation, suppression of hypermetabolic BAT would be even more pronounced. Third, further investigation is needed to quantitate the effect of the preparation on hypermetabolic BAT. Because the presence of hypermetabolic BAT is sporadic in adult humans, findings with animal models, in which hypermetabolic BAT can be induced, may be a useful platform for further study of this phenomenon.

A fourth limitation of this study was that co-localization of FDG uptake with CT fat density is a compelling but not incontrovertible argument that an area is hypermetabolic BAT [21, 22]. Pathologically proven hibernomas (benign hypermetabolic BAT tumors) have been described [23] and have exhibited considerable FDG uptake at our institution. However, to our knowledge, in no study with human subjects has it been histologically proven that fat-density FDG uptake on routine PET/CT scans represents hypermetabolic BAT. We plan to perform this analysis using needle biopsy of these presumed sites of hypermetabolic BAT in human volunteers. Fifth, we expect but have not proven that tumor uptake either is not affected by or increases with use of the high-fat diet. Tumor metabolism is dysregulated. The existence of the Warburg effect (increased and dysregulated glycolysis in neoplastic cells) [24] suggests that tumor glycolysis is independent of normal control mechanisms and unaffected by fatty acid levels. Tumor FDG uptake may be increased when hypermetabolic BAT FDG uptake is suppressed by the high-fat preparation, because the pool of FDG available for uptake would be expected to increase [7].

A timed high-fat preparation protocol minimizes FDG uptake by hypermetabolic BAT and should decrease false-positive findings on oncologic PET scans. Use of this preparation also may increase sensitivity for detection of tumor uptake by allowing a larger pool of FDG to be available.

E-Z-M, Inc. has paid royalties to the authors under a patent assigned to Beth Israel Deaconess Medical Center to produce and market Clearscan, used to prepare patients for PET at that institution.

Address correspondence to G. M. Kolodny ().

References
Previous sectionNext section
1. Sell H, Deshaies Y, Richard D. The brown adipocyte: update on its metabolic role. Int J Biochem Cell Biol 2004; 36:2098 –2104 [Google Scholar]
2. Cannon B, Nedergaard J. Brown adipose tissue: function and physiological significance Physiol Rev 2004; 84:277 –359 [Google Scholar]
3. Hany TF, Gharehpapagh E, Kamel EM, Buck A, Himms-Hagen J, von Schulthess GK. Brown adipose tissue: a factor to consider in symmetrical tracer uptake in the neck and upper chest region. Eur J Nucl Med Mol Imaging 2002;1393 –1398 [Google Scholar]
4. Cohade C, Osman M, Pannu HK, Wahl RL. Up-take in supraclavicular area fat (“USA-Fat”): description on 18F-FDG PET/CT. J Nucl Med 2003; 44:170–176 [Google Scholar]
5. Shreve PD, Anzai Y, Wahl RL. Pitfalls in oncologic diagnosis with FDG PET imaging: physiologic and benign variants. RadioGraphics 1999; 19:61–77 [Google Scholar]
6. Rousseau C, Bourbouloux E, Campion L, et al. Brown fat in breast cancer patients: analysis of serial (18)F-FDG PET/CT scans. Eur J Nucl Med Mol Imaging 2006; 33:780–791 [Google Scholar]
7. Bogsrud TV, Lowe V. Normal variants and pitfalls in whole body PET imaging with 18F FDG. Appl Radiol 2006; 35:16 –30 [Google Scholar]
8. Christensen CR, Clark PB, Morton KA. Reversal of hypermetabolic brown adipose tissue in F-18 FDG PET imaging. Clin Nucl Med 2006; 31:193 –196 [Google Scholar]
9. Tatsumi M, Engles JM, Ishimori T, Nicely O, Cohade C, Wahl RL. Intense (18)F-FDG uptake in brown fat can be reduced pharmacologically. J Nucl Med 2004; 45:1189 –1193 [Google Scholar]
10. Soderlund V, Larsson SA, Jacobsson H. Reduction of FDG uptake in brown adipose tissue in clinical patients by a single dose of propranolol. Eur J Nucl Med Mol Imaging 2007; 34:1018 –1022 [Google Scholar]
11. Gelfand MJ O'Hara SM, Curtwright LA, McLean JR. Pre-medication to block [(18)F]FDG uptake in the brown adipose tissue of pediatric and adolescent patients. Pediatr Radiol 2005; 35:984–990 [Google Scholar]
12. Baba S, Tatsumi M, Ishimori T, Lilien DL, Engles JM, Wahl RL. Effect of nicotine and ephedrine on the accumulation of 18F-FDG in brown adipose tissue. J Nucl Med 2007; 48:981–986 [Google Scholar]
13. Cinti S. The adipose organ: morphological perspectives of adipose tissues. Proc Nutr Soc 2001; 60:319–328 [Google Scholar]
14. Williams G, Kolodny GM. Suppression of myocardial 18F-FDG uptake by preparing patients with a high-fat, low-carbohydrate diet. AJR 2008; 190:466; [web]W151–W156 [Abstract] [Google Scholar]
15. Frayn KN. The glucose-fatty acid cycle: a physiological perspective. Biochem Soc Trans 2003; 31:1115 –1119 [Google Scholar]
16. Rousset S, Alves-Guerra MC, Mozo J, et al. The biology of mitochondrial uncoupling proteins. Diabetes 2004; 53[suppl 1]:S130 –S135 [Google Scholar]
17. Cooney GJ, Newsholme EA. The maximum capacity of glycolysis in brown adipose tissue and its relationship to control of the blood glucose concentration FEBS Lett 1982; 148:198–200 [Google Scholar]
18. Sidossis LS, Stuart CA, Shulman GI, Lopaschuk GD, Wolfe RR. Glucose and insulin-induced inhibition of fatty acid oxidation: the glucose-fatty acid cycle reversed. Am J Physiol 1996; 270:E733 –E738 [Google Scholar]
19. Rabelo R, Camirand A, Silva JE. 3,5-Cyclic adenosine monophosphate-response sequences of the uncoupling protein gene are sequentially recruited during darglitazone-induced brown adipocyte differentiation. Endocrinology 1997; 138:5325 –5332 [Google Scholar]
20. Valverde AM, Benito M. The brown adipose cell: a model for understanding the molecular mechanisms of insulin resistance. Acta Physiol Scand 2005; 183:59 –73 [Google Scholar]
21. Minotti AJ, Shah L, Keller K. Positron emission tomography/computed tomography fusion imaging in brown adipose tissue. Clin Nucl Med 2004; 29:5 –11 [Google Scholar]
22. Cohade C, Mourtzikos KA, Wahl RL. “USA-fat”: prevalence is related to ambient outdoor temperature-evaluation with 18F-FDG PET/CT. J Nucl Med 2003; 44:1267 –1270 [Google Scholar]
23. Subramaniam RM, Clayton AC, Karantanis D, Collins DA. Hibernoma: 18F FDG PET/CT imaging. J Thorac Oncol 2007; 2:569 –570 [Google Scholar]
24. Kelloff GJ, Hoffman JM, Johnson B, et al. Progress and promise of FDG-PET imaging for cancer patient management and oncologic drug development. Clin Cancer Res 2005; 11:2785 –2808 [Google Scholar]

Recommended Articles

Method for Decreasing Uptake of 18F-FDG by Hypermetabolic Brown Adipose Tissue on PET

Full Access,
American Journal of Roentgenology. 2009;193:359-366. 10.2214/AJR.08.2081
Abstract | Full Text | PDF (1184 KB) | PDF Plus (1013 KB) 
Full Access,
American Journal of Roentgenology. 2008;190:W151-W156. 10.2214/AJR.07.2409
Abstract | Full Text | PDF (647 KB) | PDF Plus (795 KB) 
Full Access,
American Journal of Roentgenology. 2008;190:1149-1151. 10.2214/AJR.07.3981
Citation | Full Text | PDF (466 KB) | PDF Plus (526 KB) 
Full Access, , , , , , ,
American Journal of Roentgenology. 2004;183:1127-1132. 10.2214/ajr.183.4.1831127
Abstract | Full Text | PDF (1529 KB) | PDF Plus (1646 KB) 
Full Access, , ,
American Journal of Roentgenology. 2008;190:W310-W314. 10.2214/AJR.07.2195
Abstract | Full Text | PDF (639 KB) | PDF Plus (702 KB) 
Full Access,
American Journal of Roentgenology. 2008;190:W304-W309. 10.2214/AJR.07.2928
Abstract | Full Text | PDF (1102 KB) | PDF Plus (1011 KB)