Fluorine-18–labeled sodium fluoride was first described for use in skeletal imaging techniques applied in humans in 1962, prior to the advent of PET imaging.1 The favorable imaging characteristics of 18F-NaF include a high lesion-to-normal bone concentration ratio shortly following injection, rapid plasma clearance (resulting in superior skeletal imaging with minimal soft tissue visualization), low radiation dose to the patient, good counting-rate efficiency, and rapid decay.2 Historically, the high-energy 511-keV annihilation photons produced by the positron emission decay of 18F-NaF could not be adequately imaged with standard Anger-type gamma cameras and required rectilinear scanners equipped with thick thallium–activated sodium iodide crystals.3 Difficulty in acquisition of images with the high-energy annihilation photons limited the clinical use of 18F-NaF and promoted the development of 99mTc-labeled bone imaging agents suitable for use with standard gamma cameras, which were optimized for the 140-keV photons of 99mTc. Ultimately, 99mTc-labeled diphosphonates became the standard agent for skeletal scintigraphy.3 With PET scanner development in the 1970s and the commercial availability of hybrid PET-CT instruments in the early 2000s,4 skeletal imaging techniques that used 18F-NaF–PET provided improved spatial resolution, greater sensitivity for lesion detection, and improved lesion localization on cross-sectional images, compared with 99mTc-diphosphonate skeletal scintigraphy.3,5 Although 18F-NaF has been approved by the US FDA since 1972,6 its use in human diagnostic imaging procedures has not been widespread because of lack of insurance coverage for that particular use of the radiopharmaceutical. However, the use of skeletal imaging applications in human medicine is now expected to greatly increase because reimbursement for 18F-NaF–PET scans has been approved by the Centers for Medicare and Medicaid Services as of February 2011 for sites participating in the National Oncologic Positron Emission Tomography Registry.7 In addition, sporadic global shortages of 99mTc for medical and veterinary use in the past few years, increasing availability of hybrid PET-CT instruments at or near veterinary institutions, and widespread development of commercial cyclotrons capable of creating and delivering PET isotopes to customers have all contributed to increased interest in the use of 18F-NaF as a skeletal imaging agent in veterinary medicine. To the authors' knowledge, biodistribution of 18F-NaF in clinically normal dogs has not been described. The primary goals of the study reported here were to determine the ideal interval to image acquisition after IV injection of 18F-NaF as part of an imaging protocol for a whole-body 18F-NaF PET-CT bone scanning and evaluate biodistribution of the radiopharmaceutical in clinically normal skeletally immature dogs.
Materials and Methods
This study was approved by the Colorado State University Institutional Animal Care and Use Committee. Four purpose-bred sexually intact female hound-mix dogs with a mean age of 9 months (range, 7 to 10.5 months) and mean body weight of 22.9 kg (range, 20.8 to 25.6 kg) were enrolled in the study. The dogs were determined to be healthy and free of any musculoskeletal lameness on the basis of the results of physical and lameness examinations, a CBC, serum biochemical analysis, and urinalysis.
Each dog was premedicated with a combination of atropine (0.02 mg/kg), morphine (1 mg/kg), and acepromazine maleate (0.01 mg/kg) administered SC. Anesthesia was induced with a combination of ketamine (5.6 mg/kg) and diazepam (0.28 mg/kg) administered IV and maintained with isoflurane in oxygen. The dog received positive-pressure ventilation at a volume of 10 to 15 cm H2O. Separate IV catheters were placed for fluid administration (right cephalic vein), radiopharmaceutical injection (left or right saphenous vein), and venous blood sample collection (left cephalic vein). Urine was also collected via a Foley catheter and maintained in a shielded container. In 1 dog, a gamma countera was used to monitor the blood concentration of 18F-NaF (counts/min) every 5 minutes for 200 minutes after radiopharmaceutical injection.
Image acquisition—The PET-CT examinations were performed with a PET-CT hybrid instrumentb consisting of a 16-slice, helical CT x-ray tube and detector configuration and a lutetium:yttrium orthosilicate PET detector ring. Once anesthetized, each dog was placed in sternal recumbency on the CT couch, with the pelvic limbs entering the CT and PET detectors first. The forelimbs were flexed caudally along the thoracic wall of each dog so that the forelimbs would be included within the dynamic acquisition field of view. A foam trough was used to maintain patient positioning. Once each dog was positioned, a low–radiation dose, whole-body, non–contrast medium–enhanced CT scan (120 kV; 200 mA; field of view, 600 mm; matrix, 768 × 768; slice thickness, 5 mm) was performed for attenuation correction and anatomic localization prior to the PET examination. Following the whole-body CT scan, a 2-hour, dynamic list-mode PET scan (field of view, 576 mm; matrix, 144 × 144), consisting of 24 frames at 5 min/frame and bed length of 180 mm, was initiated simultaneously with the IV bolus injection of 18F-NaF (5.18 MBq/kg [0.14 mCi/kg]). The dynamic scans were centered over the thorax so as to include the heart, aorta, liver, thoracic skeleton, and humeri.
Following the 2-hour dynamic examination of each dog, a static PET examination of the entire skeleton was obtained (field of view, 576 mm; matrix, 144 × 144). Each static examination consisted of 8 to 9 bed positions/dog (bed length, 180 mm/position) with an acquisition time of 3 min/bed, for a total scan time of 24 to 27 minutes. At the completion of the static acquisition, each dog was transported to the nuclear medicine patient ward for recovery from anesthesia. Venous catheters were removed. During recovery from anesthesia, each dog was supervised by remote monitoring to minimize personnel radiation exposure. Each dog remained in the nuclear medicine patient ward, and starting 4 hours after radiopharmaceutical injection, the surface exposure rate at the shoulder was checked. Once the surface exposure rate at the shoulder was ≤ 2 mR/h, the dog was released to the general research population wards; release occurred 4 to 8 hours after injection for all dogs.
Image analysis—Maximum intensity images were reconstructed from the static acquisition data and reviewed to assess overall image quality and pattern of radiopharmaceutical distribution. Transverse plane static images were also evaluated for image quality, pattern of radiopharmaceutical distribution, and determination of artifacts. These images were reconstructed into dorsal and sagittal planes for additional evaluation.
Quantitative analysis was performed with standard commercial software.c The dynamic 18F-NaF scans were analyzed by placing an identically sized circular ROI (10 pixels [ie, an area of 0.501 to 1.088 cm2, depending on dog size) over the ascending aorta, liver, longissimus muscular system at the level of the cranial thoracic vertebral column, second thoracic vertebral body, and humeral head bilaterally. Correct placement of the ROIs on the target organs was aided by fusing or concurrent viewing of the PET image with the anatomic CT image (Figure 1). The SUVs were calculated with a standard formula by the standard commercial softwarec on the basis of the mean tissue concentration of 18F-NaF (decay corrected) at time CPET(T) for the ROIs divided by the radioactivity injected (total activity in syringe minus residual activity remaining within the syringe after dose administration) per kilogram of body weight:
Illustration of the ROIs (indicated by circles) on transverse CT (A, C, and E) and PET (B, D, and F) images obtained from 1 of 4 female dogs used to determine the ideal interval to image acquisition after IV injection of 18F-NaF and evaluate distribution of the radiopharmaceutical in the blood pool (measured at the ascending aorta [A and B]); humeral heads, second thoracic vertebral body, and skeletal muscle (C and D); and liver (E and F) in clinically normal skeletally immature dogs.
Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1589
For SUV calculation, a residual activity of 9.25 MBq (0.25 mCi) was used on the basis of the residual activity from multiple PET radioisotope injections previously calculated at our institution. An SUV was measured for each anatomic region obtained from the ROI data in the dynamic acquisition procedures and used to generate time-activity curves. The time-activity curves were evaluated to determine the optimal interval after IV administration of 18F-NaF for acquisition of images with which to evaluate skeletal muscle and liver uptake, and to assess time from injection to blood pool clearance of the radiopharmaceutical.
Histologic examination—Approximately 1 week following the PET-CT examinations, the 4 dogs were euthanized as part of a separate research study, which did not involve the skeletal system. Euthanasia was achieved by means of IV bolus injection of sodium pentobarbital. Tissues of the second cervical vertebrae, proximal humeral physes, and distal femoral physes were collected from each dog and submitted for histologic evaluation to confirm normal bone development.
Results
SUVs for humeral head, vertebral body, aorta, liver, and skeletal muscle tissues—A time-activity curve of the mean SUV for each tissue evaluated in all dogs was created (Figure 2). A marked difference was seen between skeletal uptake of 18F-NaF and uptake in the other tissues. The humeral head was the tissue with the highest SUV; mean ± SD values were 2.39 ± 0.60 (range, 2.654 to 1.801) at 5 minutes, 7.98 ± 1.39 (range, 9.858 to 6.486) at 20 minutes, 12.74 ± 2.58 (range, 15.457 to 9.336) at 1 hour, and 15.24 ± 3.74 (range, 17.708 to 10.580) at 2 hours after injection. The anatomic area with the second highest SUV was the second thoracic vertebral body; mean values were 2.21 ± 0.29 (range, 1.819 to 2.523) at 5 minutes, 5.85 ± 0.57 (range, 6.267 to 5.025) at 20 minutes, 8.40 ± 0.90 (range, 9.65 to 4.621) at 1 hour, and 9.34 ± 0.91 (range, 10.709 to 8.859) at 2 hours after injection (Figure 3). After 2 hours, humeral heads and vertebral bodies continued to have a steady subtle increase in SUV The uptake in skeletal muscle was the lowest of uptakes among all evaluated tissues; the highest mean SUV for skeletal muscle was 0.51 ± 0.09 (range, 0.635 to 0.423) at 10 minutes following injection, which rapidly decreased to 0.21 ± 0.06 (range, 0.289 to 0.142) at 105 minutes after injection. The uptake in the liver and blood pool followed a similar pattern as did the skeletal muscle uptake but with slightly higher SUVs. In the liver, the highest mean SUV was 1.50 ± 0.62 (range, 2.242 to 0.739), which was detected at 5 minutes after injection; there was an abrupt decrease in SUV to 0.6 ± 0.36 (range, 1.069 to 0.202) at 20 minutes and 0.23 ± 0.19 (range, 0.501 to 0.18) at 60 minutes after injection. The highest mean blood pool uptake (measured at the ascending aorta) was 3.95 ± 0.61 (range, 4.651 to 3.316) at 5 minutes after injection, with an abrupt decrease in mean SUV to 1.30 ± 0.25 (range, 1.568 to 1.082) at 20 minutes and 0.52 ± 0.14 (range, 0.686 to 0.393) at 60 minutes after injection. After 60 minutes, the uptake in the liver and blood pool continued to slowly decrease. Results of blood analysis with a gamma counter in 1 dog also indicated an abrupt decrease in the amount of 18F-NaF in the circulation, with 270,991.1 counts/min at 5 minutes, 56,740.5 counts/min at 45 minutes, and 4,512.4 counts/min at 200 minutes after radiopharmaceutical injection.
Time-activity curves for mean SUV in the humeral heads (black diamonds), second thoracic vertebral body (black and white hexagons), blood pool (measured at the ascending aorta [white circles]), liver (black and white squares), and skeletal muscle (white hexagon with cross) in 4 skeletally immature female dogs following IV administration of 18F-NaF (0.14 mCi/kg).
Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1589
Time-activity curves for mean SUV in the humeral heads (A), second thoracic vertebral body (B), blood pool (measured at the ascending aorta [C]), liver (D), and skeletal muscle (E) in each of 4 skeletally immature female dogs following IV administration of 18F-NaF (0.14 mCi/kg). Humeral heads (A) had the highest SUV, followed by that for the vertebral bodies (B); uptake in skeletal muscle (E) was the lowest of all evaluated tissues. The SUVs in the blood pool (C) and liver (D) were similar but slightly higher than findings for skeletal muscle (E).
Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1589
The 2-hour postinjection static images including the reconstructed maximum intensity images revealed a characteristic distribution of 18F-NaF in bone and an excellent bone-to-soft tissue ratio with only bone visible without soft tissue background (Figure 4). Areas of increased uptake were localized to areas of open centers of ossification and growth plates, including the epiphyseal regions of long bones, vertebral and sternebral body end plates, sacroiliac joint, pubic symphysis, femoral heads, greater trochanters, and costochondral junctions. Results of a CBC, serum biochemical analysis, urinalysis, and histologic examination of the second thoracic vertebral body, proximal humeral physes, and distal femoral physes were considered normal for all 4 dogs.
Representative maximum intensity 3-D whole-body projection from the lateral (A) and dorsoventral (B) perspectives reconstructed from a PET scan of a skeletally immature female dog following IV administration of 18F-NaF (0.14 mCi/kg). Notice the excellent radioisotope localization to bone as well as the level of skeletal anatomic detail achieved by use of this technique.
Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1589
Discussion
In the human medical literature, it has been reported that the radiopharmaceutical 18F-NaF has the desirable characteristics of high and rapid bone uptake accompanied by very rapid blood clearance, which results in a high bone-to-background ratio in a short time.3,8 The mechanism of uptake of 18F-NaF is similar to other bone-seeking radiopharmaceuticals, with accumulation depending on regional blood flow and regional osteoblastic activity.7 The fluoride of 18F-NaF is analogous to chloride and preferentially deposits at sites of high bone turnover and remodeling by chemisorption onto bone surfaces, exchanging with hydroxyl groups in hydroxyapatite crystals of bone to form fluoroapatite.8–10 After IV administration in humans, 18F-NaF is rapidly cleared from the plasma in a biexponential manner. The first phase has a half-life of 0.4 hours, and the second phase has a half-life of 2.6 hours. Essentially, all of the 18F-NaF that is delivered to bone by the blood circulation is retained in the bone.3 In the present study, the rapid blood clearance of 18F-NaF was also demonstrated via gamma counter analysis in 1 dog. In addition, the dynamic PET-CT image acquisitions in all dogs provided evidence of the same abrupt decrease of 18F-NaF in the blood pool (measured at the ascending aorta) soon after injection of the radiopharmaceutical, thereby confirming the rapid blood clearance of 18F-NaF in dogs.
As with other bone-seeking radiopharmaceuticals, a certain amount of time must elapse following administration prior to image acquisition to allow adequate accumulation of the isotope in areas of osteoblastic activity and for adequate soft tissue clearance of the radiopharmaceutical to occur. Compared with 99mTc–methyl diphosphonate, 18F-NaF provides better accuracy for quantitative bone evaluations because it has a smaller molecular weight and higher capillary permeability and its kinetics are not affected by protein binding.10 The half-life of 18F-NaF (110 minutes) is shorter than that for 99mTc (6 hours), allowing more rapid clearance of radioactive material from patients. The renal clearance of 18F-NaF is dependent on urine flow rate; therefore, good hydration of patients is important. At high urine flows (≥ 5 mL/min), fluoride renal clearance is 60% to 90% of the glomerular filtration rate; however, at flows of < 1 mL/min), renal clearance may be as low as 5% of the glomerular filtration rate.10 Based on the time-activity curves in the study reported here, quality images characterized by high image contrast, a good signal-to-noise ratio, and a good bone-to-soft tissue ratio should be obtainable at approximately 50 minutes after injection of the radiopharmaceutical. This uptake time is similar to the times reported for skeletal imaging procedures in humans.3,5,9 The time-activity curves in our study also indicated that after 50 minutes, the bone-to-soft tissue ratio continued to slightly increase; however, this subtle change in ratio was not expected to substantially further improve image quality because approximately 1 hour after administration of 18F-NaF, only approximately 10% of the injected dose remains in the blood.3 Although a minimal image quality improvement might still be gained by waiting > 50 minutes for image acquisition, this potential improvement will likely not be enough to justify further prolongation of the postinjection period.
Compared with standard planar skeletal scintigraphy involving the use of 99mTc radioisotopes, the use of PET technology in conjunction with administration of 18F-NaF provides better spatial resolution and greater sensitivity for lesion detection and has the added advantage of cross-sectional (tomographic) image acquisition and quantitative analysis by SUV calculation.3,9 Because of these useful characteristics, and in combination with the anatomic and structural information derived from the CT images, 18F-NaF PET-CT imaging has many potential advantages over conventional 99mTc bone scintigraphy for the detection of skeletal abnormalities.9 One limitation of 18F-NaF scintigraphy is the high energy of the 511-keV annihilation photons produced by decay of the radioactive fluorine.3 To obtain images from these high-energy photons, expensive and specialized equipment that currently has low availability in veterinary medicine has to be used; moreover, high-level shielding of imaging suites and specific, shielded patient waiting or housing areas are required. One additional disadvantage of PET-CT imaging procedures in veterinary patients is the need for anesthesia. The use of sedation, similar to protocols used in some veterinary CT studies, would not be feasible for PET-CT imaging procedures in animals because of the duration of PET image acquisition and the need for complete patient immobility to obtain the desired image registration match between PET and CT scans. With a sedated patient, there would also be unnecessary excess radiation exposure of personnel as a result of increased patient contact for observation during examinations.
Sodium fluoride F 18 has many potential applications in veterinary medicine, including evaluation of skeletal metastases, skeletal kinetics, fracture healing, and bone viability. The behavior of 18F-NaF and 99mTc-labeled bone-seeking agents is very similar, although 18F-NaF has more rapid blood pool and soft tissue clearance. Extraosseous uptake of 99mTc-based radiopharmaceuticals has been useful in the diagnosis of a variety of conditions in humans and other animals, including neoplastic, hormonal, inflammatory, ischemic, traumatic, and excretory diseases.8,11 The behavior of 18F-NaF in human and veterinary patients with soft tissue lesions has not been studied, to the authors' knowledge, and its potential applications in the diagnosis of these conditions are yet to be determined.
Skeletal distribution of 18F-NaF in 4 clinically normal skeletally immature dogs was determined in the present study, and these data could serve as a baseline for future studies of 18F-NaF in veterinary medicine, particularly with the increasing use of PET-CT at veterinary institutions. However, if skeletally mature dogs are used, one should be aware that the distribution of the radiopharmaceutical may differ from that determined in the present study, especially in areas of increased bone turnover, such as growth plates and centers of ossification.
ABBREVIATIONS
| 18F-NaF | Sodium fluoride F 18 |
| 99mTc | Technetium Tc 99 metastable |
| PET | Positron emission tomography |
| ROI | Region of interest |
| SUV | Standardized uptake value |
Wizard, PerkinElmer, Life Sciences, Turku, Finland.
Gemini TF Big Bore, Philips Healthcare, Andover, Mass.
Extended Brilliance, Fusion Viewer, Philips Healthcare, Andover, Mass.
References
- 1.↑
Blau MNagler WBender MA. Fluorine-18: a new isotope for bone scanning. J Nucl Med 1962; 3:332–334.
- 2.↑
Weber DAGreenberg EJDimich A, et al. Kinetics of radionuclides used for bone studies. J Nucl Med 1969; 10:8–17.
- 3.↑
Grant FDFahey FHPackard AB, et al. Skeletal PET with 18F-fluoride: applying new technology to an old tracer. J Nucl Med 2008; 49:68–78.
- 4.↑
Workman RB JrColeman RE. Fundamentals of PET and PET/CT imaging. In: Workman RB JrColeman RE, eds. PET/CT essentials for clinical practice. New York: Springer Science+Business Media LLC, 2006;1–22.
- 5.
Hoh CKHawkins RADahlbom M, et al. Whole body skeletal imaging with [18F]fluoride ion and PET. J Comput Assist Tomogr 1993; 17:34–41.
- 6.↑
Bridges RLWiley CRChristian JC, et al. An introduction to Na(18)F bone scintigraphy: basic principles, advanced imaging concepts, and case examples. J Nucl Med Technol 2007; 35:64–76.
- 7.↑
American College of Radiology. National oncologic PET registry enables expanded coverage for F-18 sodium fluoride PET scans. Available at: www.acr.org/SecondaryMainMenuCategories/NewsPublications/FeaturedCategories/CurrentACRNews/Expanded-Coverage-PET-Scans.aspx. Accessed Feb 15, 2011.
- 8.
Even-Sapir EMishani EFlusser G, et al. 18F-Fluoride positron emission tomography and positron emission tomography/computed tomography. Semin Nucl Med 2007; 37:462–469.
- 9.↑
Cook JRFogelman IIsrael O. PET imaging of the skeleton. In: Valk PEDelbeke DBailey DL, et al, eds. Positron emission tomography: clinical practice. London: Springer, 2006;317–335.
- 10.↑
Blake GMPark-Holohan SJCook GJ, et al. Quantitative studies of bone with the use of 18F-fluoride and 99mTc-methylene diphosphonate. Semin Nucl Med 2001; 31:28–49.
- 11.
Peller PJHo VBKransdorf MJ. Extraosseous Tc-99m MDP uptake: a pathophysiologic approach. Radiographics 1993; 13:715–734.
