OtherCONTINUING EDUCATION

Skeletal PET with 18F-Fluoride: Applying New Technology to an Old Tracer

Frederick D. Grant, Frederic H. Fahey, Alan B. Packard, Royal T. Davis, Abass Alavi and S. Ted Treves
Journal of Nuclear Medicine January 2008, 49 (1) 68-78; DOI: https://doi.org/10.2967/jnumed.106.037200

Abstract

Although 18F-labeled NaF was the first widely used agent for skeletal scintigraphy, it quickly fell into disuse after the introduction of 99mTc-labeled bone-imaging agents. Recent comparative studies have demonstrated that 18F-fluoride PET is more accurate than 99mTc-diphosphonate SPECT for identifying both malignant and benign lesions of the skeleton. Combining 18F-fluoride PET with other imaging, such as CT, can improve the specificity and overall accuracy of skeletal 18F-fluoride PET and probably will become the routine clinical practice for 18F-fluoride PET. Although 18F-labeled NaF and 99mTc-diphosphonate have a similar patient dosimetry, 18F-fluoride PET offers shorter study times (typically less than 1 h), resulting in a more efficient workflow, improved patient convenience, and faster turnarounds of reports to the referring physicians. With the widespread availability of PET scanners and the improved logistics for the delivery of 18F radiopharmaceuticals, prior limitations to the routine use of 18F-fluoride bone imaging have largely been overcome. The favorable imaging performance and the clinical utility of 18F-fluoride PET, compared with 99mTc-diphosphonate scintigraphy, support the reconsideration of 18F-fluoride as a routine bone-imaging agent.

Several decades before the introduction of modern PET systems, 18F-labeled NaF was recognized as an excellent radiopharmaceutical for skeletal imaging (1). 18F-Fluoride 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. High-quality images of the skeleton can be obtained less than an hour after the intravenous administration of 18F-labeled NaF. 18F-labeled NaF became widely used for skeletal scintigraphy after its introduction by Blau and others in the early 1960s (2) and was approved for clinical use by the U.S. Food and Drug Administration in 1972. One limitation of 18F-fluoride scintigraphy was the relatively high energy of the 511-keV annihilation photons produced by decay of 18F. This high energy necessitated imaging with rectilinear scanners equipped with relatively thick NaI(Tl) crystals and precluded the use of Anger-type γ-cameras. This technical limitation, combined with the widespread availability of 99Mo/99mTc generators, encouraged the development of 99mTc-labeled bone agents.

In the early 1970s, 99mTc-labeled polyphosphates and then 99mTc-labeled pyrophosphate were introduced as bone-imaging agents. With readily available 99mTc, bone scintigraphy quickly became one of the most commonly performed nuclear medicine imaging procedures (3). When it became apparent that pyrophosphate impurities or degradation products were responsible for most of the bone-imaging properties of 99mTc-labeled polyphosphates, 99mTc-polyphosphates were abandoned in favor of 99mTc-pyrophosphate (4). However, skeletal imaging with 99mTc-pyrophosphate was limited by prolonged clearance from the circulation. During this same period, 99mTc-labeled diphosphonates were introduced for skeletal scintigraphy. These compounds demonstrated higher skeletal uptake, faster blood-pool clearance, and better in vivo stability than did either polyphosphates or pyrophosphate. With the successful development of kit-based 99mTc-diphosphonate radiopharmaceuticals and the increased availability of Anger-type γ-cameras, 99mTc-diphosphonates, particularly 99mTc-methylene diphosphonate (MDP), which demonstrated faster blood-pool clearance than did most other 99mTc-labeled diphosphonates, became adopted as the standard agent for skeletal scintigraphy (3,4). The precipitous decline in the use of 18F-labeled NaF for skeletal scintigraphy did not, therefore, reflect limitations of 18F-labeled NaF as a tracer per se, but instead was the result of, first, the difficulty in imaging 511-keV photons on a system optimized for the 140-keV photons of 99mTc and, second, the logistic challenges in the production and efficient delivery of a radioisotope with a physical half-life of 110 min.

In the early 1990s, based on the favorable skeletal kinetics of 18F-fluoride (5), Phelps et al. used 18F-fluoride PET as a model for the development of whole-body PET (6). Current PET scanners have higher spatial resolution and substantially greater sensitivity than do conventional γ-cameras, resulting in higher image quality for PET (Fig. 1) than for planar scintigraphy or SPECT. Driven by the demand for 18F-FDG, an efficient commercial system now exists for the production and delivery of 18F-labeled NaF. Therefore, because prior technical and logistic limitations to the routine use of 18F-fluoride for bone imaging were no longer present, the increasing availability of PET systems renewed interest in using 18F-labeled NaF as a radiotracer for skeletal imaging.

FIGURE 1. 
FIGURE 1. 

Normal 18F-fluoride PET skeletal findings for patients aged 5 y (A), 11 y (B), 15 y (C), 19 y (D), and 30 y (E). Pattern of 18F− uptake in skeleton is similar to pattern seen with more familiar 99mTc-labeled bisphosphonate bone scans and illustrates changes that occur with maturation of skeleton. Compared with 99mTc-MDP SPECT, 18F-fluoride bone PET provides higher-quality images, better ratios of bone uptake to soft-tissue uptake, and shorter studies. MIP = maximum-intensity projection.

In the past decade, the clinical utility of 18F-fluoride PET bone scans has been demonstrated by numerous studies, which will be discussed in this review. In these studies, direct comparison with 99mTc-MDP bone scans has shown 18F-fluoride PET to have a higher diagnostic accuracy in the assessment of malignant and benign skeletal disease. The widespread readoption of 18F-labeled NaF as a bone-imaging agent, however, has been limited by longstanding familiarity with 99mTc-diphosphonate scintigraphy and, possibly, by concerns about insurance reimbursement for 18F-fluoride PET.

PRODUCTION AND PHARMACOLOGY OF 18F

Historically, one reason for the early interest in 18F-fluoride for bone imaging, other than its high affinity for bone, was that it is relatively easy to produce in high specific activity in a nuclear reactor using a 2-step reaction starting with an Li2CO3 target.MathMathThe high-energy tritons produced by fission of 6Li react with the 16O from the Math in the LiCO3 target to produce 18F (7). No other fluorine isotopes are produced, so the material is carrier-free. The reaction does, however, produce a significant amount of 3H that must be removed before the tracer can safely be administered to patients. These large amounts of 3H also present a significant radioactive waste disposal problem.

Currently, nearly all 18F used for clinical applications is produced in a cyclotron using a 1-step reaction with isotopically enriched (≥95%) 18OH2 (water) as the target (8):MathSeveral curies of 18F can be produced using short irradiation times (2 h) with beam currents that are achievable with clinical cyclotrons. The 18F that is produced in this reaction is collected as 18F− by passage of the irradiated water through an ion-exchange column. If the 18F− is to be used for bone scanning, no further chemical processing is required.

After intravenous administration, 18F-fluoride is rapidly cleared from the plasma in a biexponential manner. The first phase has a half-life of 0.4 h, and the second phase has a half-life of 2.6 h (9). Essentially all the 18F-fluoride that is delivered to bone by the blood is retained in the bone (10). Tracer retention by the bone is a 2-phase process (11). In the first phase, the 18F− ion exchanges for an OH− ion on the surface of the hydroxyapatite matrix of bone. In the second phase, the 18F− ion migrates into the crystalline matrix of bone, where it is retained until the bone is remodeled. One hour after administration of 18F-labeled NaF, only about 10% of the injected dose remains in the blood (1).

Approximately 30% of the injected dose of 18F-fluoride is sequestered within circulating red blood cells. However, 18F-fluoride is freely diffusible from the red cells to the bone surface; moreover, red blood cell uptake does not appear to interfere with uptake of the tracer by bone (12). The total uptake of 18F-fluoride by the bone is similar to that of 99mTc-MDP, at approximately 50% of the injected dose (13). There is minimal binding of 18F-fluoride by serum proteins (12). This is an important difference between 18F-fluoride and 99mTc-MDP and other 99mTc-diphosphonate bone agents, all of which show significant protein binding. Approximately 30% of 99mTc-MDP is protein-bound immediately after injection; this fraction increases to approximately 70% by 24 h after injection (14). The non–protein-bound fraction of 99mTc-MDP is rapidly cleared from the blood with a half-life similar to that of 18F-fluoride, but the protein-bound fraction is cleared much more slowly (15). Hence, it is necessary to wait 3–4 h after injection of 99mTc-MDP before imaging. By comparison, imaging can be performed less than 1 h after 18F-labeled NaF administration.

CLINICAL USE OF 18F-FLUORIDE: ONCOLOGY

One of the first well-described applications of 18F-fluoride PET was the imaging of primary bone tumors. Using projection and tomographic images, both malignant and benign skeletal lesions were identified by 18F-fluoride PET in 18 subjects (6,16). Subsequent clinical use has demonstrated the utility of 18F-fluoride PET in detecting primary bone tumors (Fig. 2).

FIGURE 2. 
FIGURE 2. 

In 8-y-old patient with Ewing's sarcoma of right distal fibula, 18F-fluoride PET projection image shows extent of primary tumor and absence of skeletal metastases.

18F-Fluoride PET also has been used to identify skeletal metastasis in patients with a range of primary tumors. In a study of 5 breast cancer patients with multiple skeletal metastases, increased 18F-fluoride uptake was seen both in lesions with sclerotic characteristics on CT and in lytic lesions, but there was no apparent correlation between the intensity of uptake and the amount of bone matrix seen on CT. Lesions measuring less than 3 mm on CT had limited detectability on 18F-fluoride PET. No formal comparison was made between 99mTc bone scintigraphy and 18F-fluoride PET (17).

In 2 reports, Schirrmeister et al. compared 18F-fluoride PET and planar 99mTc-MDP scintigraphy in the detection of skeletal metastases in patients with a variety of solid tumors (18,19). Lesions were characterized as benign or malignant based on appearance on bone scans; additional imaging, including CT, MRI, and 131I scintigraphy; and clinical follow-up. In 44 patients with prostate, lung, or thyroid cancer, 18F-fluoride PET detected 96 metastatic lesions in 15 patients, whereas 99mTc-MDP planar scintigraphy detected only 46 metastases (18). Although 18F-fluoride PET detected all the lesions identified by 99mTc-MDP planar scintigraphy, 99mTc-MDP found only 40% of spine metastases and 82.8% of lesions identified as metastatic in the skull, chest, and extremities. In 34 patients with known or suspected metastatic breast cancer, 18F-fluoride PET detected 64 metastatic lesions in 17 patients, whereas 99mTc-MDP scintigraphy correctly identified only 29 metastases in 11 patients (19). One lesion that was later shown to be metastatic was interpreted as equivocal on 18F-fluoride PET, whereas 14 metastatic lesions were characterized as equivocal or benign by 99mTc-MDP scintigraphy. Tc-MDP SPECT was performed on a subset of 12 patients but did not identify any additional metastatic lesions. Clinical management was changed in 4 patients in whom previously undetected bone metastases were found by 18F-fluoride PET (19). Because 18F-fluoride PET was compared primarily to whole-body 99mTc-MDP planar imaging, these 2 studies could not establish whether the increased sensitivity of 18F-fluoride PET reflected increased sensitivity of 18F-fluoride as a radiotracer or the improved performance of tomographic imaging.

When 18F-fluoride PET was prospectively compared with 99mTc-MDP planar scintigraphy and SPECT in a cohort of 53 patients with small cell or non–small cell lung cancer and without previously known metastatic disease, skeletal metastases were identified in only 12 individuals; in the other 41 patients, no skeletal metastases were identified by any of the bone scans or by MRI of the entire spine (20). For the 12 patients who did have skeletal metastases, the combination of 99mTc-MDP planar images and SPECT outperformed 99mTc-MDP planar imaging alone. 99mTc-MDP planar scintigraphy identified metastatic disease in only 5 patients, whereas the addition of 99mTc-MDP SPECT identified vertebral metastases in an additional 5 patients. For 2 patients in whom the combination of 99mTc-MDP planar scintigraphy and SPECT was equivocal or negative, 18F-fluoride PET correctly identified metastatic disease, and in 1 patient, these findings resulted in a change in clinical management. In no patients did 99mTc-MDP identify metastases that were not seen on 18F-fluoride PET.

In another prospective study of an additional 103 patients at the same institution with a new diagnosis of either non–small cell (73 patients) or small cell (30 patients) lung cancer, skeletal metastases were found by MRI in 33 patients (21). 18F-Fluoride PET identified metastatic disease in 31 of these patients, 2 more patients than were identified by 99mTc-MDP planar imaging and SPECT. Two patients with negative findings on 18F-fluoride PET were found to have metastatic disease by MRI of the spine, but in another 2 patients with negative findings on 18F-fluoride PET, MRI findings that initially suggested metastatic disease were later determined to be benign—by autopsy in one and by clinical follow-up in the other. Compared with 18F-fluoride PET, 99mTc-MDP SPECT underestimated the extent of disease in 16 patients, which was nearly half the total number of patients with skeletal metastases. As a result of 18F-fluoride PET findings that were not identified on 99mTc-MDP SPECT, the clinical management of 2 patients was changed. 18F-Fluoride PET also demonstrated fewer indeterminate lesions than did 99mTc-MDP SPECT. Only 3 indeterminate lesions, 1 of which represented skeletal metastatic disease, were found on 18F-fluoride PET, compared with 13 indeterminate lesions, 3 of which were subsequently identified as skeletal metastases, on 99mTc-MDP SPECT. As an additional part of this study, 18F-FDG PET was performed on 41 patients and detected skeletal metastases in 8 of 10 patients in whom skeletal metastases previously had been identified by 18F-fluoride PET. The relative roles of 18F-FDG and 18F-fluoride in imaging skeletal metastases must be better defined, because little has been published comparing 18F-labeled NaF and 18F-FDG or other PET radiopharmaceuticals. Hoegerle has proposed the combined administration of 18F-FDG and 18F-labeled NaF, but this approach has not been widely adopted (22). In a report of 20 patients with a wide range of primary tumors, many skeletal metastases were detected by both radiopharmaceuticals (23). 18F-FDG was more likely to detect bone marrow metastases or small osteolytic lesions, presumably lesions with little or no increase in cortical bone turnover. 18F-labeled NaF was more likely to detect skeletal metastases of tumors that typically have low FDG avidity, such as thyroid cancer or renal cell cancer (Fig. 3).

FIGURE 3. 
FIGURE 3. 

Compared with 99mTc-MDP scintigraphy (left, arrows indicate sites of thoracic and lumbar spine disease), both 18F-FDG PET (middle) and 18F-fluoride PET (right) show higher sensitivity for detecting skeletal lesions. Compared with 18F-FDG PET, 18F-fluoride PET demonstrates minimal soft-tissue uptake, which increases sensitivity for detecting bone lesions that are adjacent to sites of physiologic 18F-FDG uptake or sites of 18F-FDG–avid soft-tissue disease (Reprinted with permission of (23).).

The rapid assimilation of PET/CT only a few years after the introduction of 18F-FDG PET into routine clinical practice means that hybrid imaging is now also widely available for 18F-fluoride PET. Two publications from the same institution compared the accuracy of 18F-fluoride PET and 18F-fluoride PET/CT in detecting skeletal metastases (24,25). In 44 patients with a variety of primary tumors, 18F-fluoride PET/CT was used to evaluate skeletal pain in the absence of findings on analysis of 198 of the lesions by 99mTc-MDP scintigraphy. One hundred eleven of 212 skeletal lesions were characterized as metastatic based on analysis of 198 of the lesions by histopathology (9), contemporaneous CT or MRI (64), or imaging and clinical follow-up (125). Nearly all 18F-fluoride–avid lesions (85 of 89) characterized as benign had been identified as benign on the initial CT scan, demonstrating a higher specificity for 18F-fluoride PET/CT than for 18F-fluoride PET. Among the 18 patients not diagnosed with skeletal metastases, 16 had at least 1 focus of 18F-fluoride uptake that corresponded with a benign finding on CT (24). One issue not addressed in this study, however, is the likelihood that patients referred specifically for skeletal symptoms may be more likely to have an increased incidence of benign 18F-fluoride–avid skeletal findings.

In a subsequent prospective study, planar and SPECT 99mTc-MDP bone scans, 18F-fluoride PET, and 18F-fluoride PET/CT were performed on 44 patients with high-risk prostate cancer, and 23 patients were characterized as having metastatic disease (Fig. 4) (25). In 20 patients, 18F-fluoride PET/CT findings were used as the final determinant of whether a lesion was metastatic or benign. In 3 patients with 18F-fluoride–avid lesions, but nondiagnostic CT and MRI findings, bone biopsy (1 patient) or progression of disease on subsequent imaging (2 patients) was used to establish the diagnosis of metastases. The 21 patients in whom metastatic disease was not diagnosed on 18F-fluoride PET/CT had no clinical or imaging evidence of metastases during at least 6 mo of follow-up. As was the case in prior studies, 18F-fluoride PET was more sensitive in detecting skeletal metastases than was planar 99mTc-MDP scintigraphy either alone or in combination with 99mTc-MDP SPECT. 18F-Fluoride PET detected skeletal metastases in all 23 patients, whereas 99mTc-MDP imaging detected lesions in only 18 patients. In a subset of 24 patients (13 with metastases on PET/CT and 11 without metastatic disease), 99mTc-MDP SPECT was performed of the whole body and not just 1 field of view. Whole-body 99mTc-MDP SPECT detected metastatic disease in 12 of 13 patients, whereas 18F-fluoride PET identified metastases in all 13 patients. However, in the absence of correlative CT scans, 18F-fluoride PET and combined planar imaging and SPECT with 99mTc-MDP had similar specificities. No assessment was made as to whether hybrid 18F-fluoride PET/CT was more accurate than a correlation of separately obtained 18F-fluoride PET and CT studies in identifying skeletal metastases. However, as PET/CT scanners replace stand-alone PET systems, this question may become less of an issue.

FIGURE 4. 
FIGURE 4. 

From left to right: posterior and anterior 99mTc-MDP planar scintigraphy, 99mTc-MDP multiple-field-of-view SPECT, and 18F-fluoride PET of 82-y-old patient with numerous bone metastases. As in this patient, more lesions are typically detected by SPECT than by planar imaging, and 18F-fluoride PET detects more lesions than does SPECT.

18F-Fluoride uptake also parallels 99mTc-MDP uptake in other clinical oncologic disorders. Soft-tissue uptake interpreted as extraosseous ossification has been detected by 18F-fluoride PET, including presumed pulmonary metastases of osteosarcoma and neuroblastoma (Fig. 5) (6,16). A recent case report describes a posttreatment flare phenomenon characterized by transient 18F-fluoride uptake at a site of decreased 18F-FDG uptake after a recent change in breast cancer therapy (26). As the use of 18F-fluoride PET in oncology patients becomes more common, much of the clinical experience acquired with 99mTc-MDP scintigraphy should prove applicable to 18F-fluoride PET.

FIGURE 5. 
FIGURE 5. 

In 2-y-old girl with stage IV neuroblastoma, uptake in multiple soft-tissue tumors on 18F-fluoride PET bone scan demonstrates 18F-fluoride avidity for sites of soft-tissue calcification. A = anterior; R = right.

CLINICAL USE OF 18F-FLUORIDE: BENIGN DISEASES OF THE SKELETON

Skeletal scintigraphy has become an important part of evaluating back pain in children and adolescents, particularly in young athletes (27). Two recent reports have described the utility of 18F-fluoride PET in the evaluation of these patients (28,29). In one report, 18F-fluoride PET/CT was performed on 15 young patients (aged 9–19 y) with back pain (28). Nine of the 15 patients had abnormal 18F-fluoride uptake in the spine or pelvis that correlated with abnormal CT findings. Two patients, found on CT to have herniated disks but no osseous abnormalities, did not demonstrate corresponding 18F-fluoride uptake. No sites of abnormal 18F-fluoride uptake without a corresponding CT finding were identified. In another report (Figs. 6–8⇓⇓), 94 patients (aged 4–26 y) with back pain were evaluated with 18F-fluoride PET (29). Abnormal focal 18F-fluoride uptake was identified in the spines of 52 of these patients, with 15 patients having more than one site of abnormal uptake. Additional correlative imaging, such as CT, was not performed in the absence of other indications, but the 18F-fluoride PET findings were used by the referring physicians to guide patient treatment. More recently, the use of PET/CT image fusion has improved localization and characterization of lesions identified on bone 18F-fluoride PET (Fig. 8).

FIGURE 6. 
FIGURE 6. 

In 18-y-old female marathon runner with severe lower leg and foot pain, projection PET image shows increased 18F-fluoride uptake in tibiae and both feet, indicating multiple sites of stress injury.

FIGURE 7. 
FIGURE 7. 

(A) In 12-y-old female gymnast with lower back pain, 18F-fluoride PET image shows increased focal uptake indicating stress changes in posterior elements of L5 vertebra (arrow). (B) In 21-y-old runner, 18F-fluoride PET image shows stress changes in left sacroiliac region (arrow).

FIGURE 8. 
FIGURE 8. 

(A) Fusion of 18F-fluoride PET and CT images of 17-y-old male athlete with back pain that worsened on hyperextension. 18F-Fluoride PET had identified focally increased tracer uptake in region of right pars interarticularis of L5 vertebra. Fusion image demonstrates correlation of site of 18F− uptake with nondisplaced fracture of pars that was identified by CT. (B) From top to bottom, sagittal 18F-fluoride PET, CT, and fusion images of 15-y-old female athlete with back pain after landing from high jump. Increased 18F-fluoride uptake and deformity on PET correlates to wedge compression fracture of L3 vertebra (arrows) identified on CT.

As with skeletal SPECT, 18F-fluoride PET is a sensitive method for the detection of focal changes in bones secondary to stress caused by intense sports activity. In our experience, only about half of patients who showed focal 18F-fluoride uptake in a posterior element of the lumbar spine had accompanying CT evidence of spondylolysis. Of cases in which 18F-fluoride PET findings were normal, CT confirmed the absence of abnormality in two thirds and detected pars fracture of indeterminate age in one third. The focal increase of tracer uptake in a pars interarticularis is not necessarily associated with fracture; in some cases, the increase might represent only increased bone turnover due to stress. Depending on the age of the lesion, a pars fracture may or may not show an associated focal increase of 18F-fluoride uptake. Early or active lesions will show increased tracer uptake. However, older, well-established pars fractures, presumably with little active healing or bone remodeling, may not necessarily show increased tracer uptake in the region of interest.

Skeletal 18F-fluoride PET also has been used to predict bone viability after trauma or reconstructive surgery. In 5 patients with hip fracture and clinical concern about osteonecrosis, dynamic 18F-fluoride PET was used to measure bone blood flow and assess the extent of viable bone in the femoral head (30). Either impaired blood flow or decreased 18F-fluoride influx predicted an eventual need for joint replacement surgery. In 10 patients who had undergone resurfacing arthroplasty of the hip, 18F-fluoride PET was used to demonstrate normal 18F-fluoride uptake in bone underlying the resurfacing prosthesis (31). In patients undergoing hip revision arthroplasty for prosthesis loosening, allogenic bone grafts may be required to stabilize the hip joint. Two reports have described the use of 18F-fluoride PET to assess bone graft viability. In 16 patients, dynamic 18F-fluoride PET was performed either early (3–6 wk) or late (9 mo–5 y) after surgery (32). Early studies demonstrated similar uptake in both allograft bone and adjacent cortical bone, whereas later studies showed less activity in allograft bone than in cortical bone. Another report describes 5 patients on whom dynamic 18F-fluoride PET was performed both early (1–8 d) and late (1 y) after surgery (33). Three of the 5 patients also were studied 4 mo after surgery.

In certain clinical settings, one potential limitation of 18F-fluoride PET is the inability to yield an equivalent to a 3-phase bone scan. 18F-Fluoride would not be expected to accumulate in acute inflammatory processes in soft tissue and so will not be useful to image processes that can be identified on tissue-phase images of 99mTc-MDP scintigraphy. Concurrent or sequential use of 18F-labeled NaF and 18F-FDG may be an approach to assessing soft tissue adjacent to a skeletal lesion. It is possible that 99mTc-diphosphonates will continue to have a role in clinical situations requiring a 3-phase bone scan.

Several investigators have used 18F-fluoride skeletal PET to quantitatively assess bone turnover. Quantitative 18F-fluoride PET provides a noninvasive measure of bone turnover that correlates with bone histomorphometry (34,35). The use of quantitative 18F-fluoride PET has been demonstrated in patients with renal osteodystrophy (34), postmenopausal osteoporosis (36), and Paget's disease (37). For example, in a study of 11 patients with renal osteodystrophy, the calculated rate constant of net incorporation of 18F-fluoride into bone was similar when calculated by either a 3-compartment model or Patlak analysis and correlated with serum markers of bone turnover (e.g., alkaline phosphatase) and PTH level (34). The rate of fluoride incorporation into bone also correlated with histomorphometric indices of bone turnover in iliac crest biopsies and was higher in patients with high-turnover osteodystrophy than in patients with low-turnover bone disease or in healthy subjects. Although useful as a research tool to help improve the understanding of bone metabolism, 18F-fluoride PET has not entered routine clinical practice for the diagnosis and assessment of metabolic bone disease or osteoporosis.

DOSIMETRY

Several factors affect the radiation dose of 18F relative to that of single-photon emitters (such as 99mTc). With a positron emitter such as 18F, energy is delivered by the positron itself (mean energy, 250 keV) and by the two 511-keV annihilation photons, whereas 99mTc emits a single 140-keV γ-ray, as well as conversion electrons in low abundance. These differences affect the relative internal dosimetry of 18F and 99mTc. The 18F positron will deposit essentially all its kinetic energy in the source organ, whereas the different energies of 511-keV and 140-keV photons result in different patterns of internal radiation dose. The soft-tissue half-value layers for the 511- and 140-keV photons are 7.3 and 4.6 cm, respectively, so that 511-keV photons can deliver their energy to organs distant from the source organ, whereas the 140-keV photons will deliver more of their energy to organs near the source organ. On the other hand, the half-life for 18F is 110 min, compared with 6 h for 99mTc, leading to a shorter exposure period and, in turn, to a reduced radiation dose for 18F.

With consideration of all these factors, the radiation dosimetry for both 18F-labeled NaF and 99mTc-MDP was calculated using the data provided by reports 53 and 80, respectively, of the International Commission on Radiological Protection (ICRP) (38,39). The effective doses per unit of administered activity for 18F-labeled NaF and 99mTc-MDP were calculated, as was the radiation dose to several individual organs (Table 1). Also listed in this table are the absolute doses for patients of different sizes for injected activities of 2.11 and 7.40 MBq/kg for 18F-labeled NaF and 99mTc-MDP, respectively. At the prescribed administered activities, the effective dose for 18F-labeled NaF and 99mTc-MDP are similar for most patients (4.0 and 3.0 mSv, respectively, for a 70-kg patient). For patients weighing less than 20 kg, the effective dose is relatively less with 99mTc-MDP (2.0 mSv) than with 18F-labeled NaF (approximately 3.5 mSv). The bone surface dose is higher for 99mTc-MDP relative to that for 18F-labeled NaF (32.6 and 5.9 mGy, respectively, for a 70-kg patient), whereas the dose to the bladder wall is slightly higher for 18F-labeled NaF than for 99mTc-MDP (32.6 and 24.9 mGy, respectively, for a 70-kg patient). Therefore, there is little overall difference in the estimated radiation dose for 18F-labeled NaF PET and 99mTc-MDP scintigraphy.

View this table:
TABLE 1

Radiation Dosimetry of 99mTc-MDP Scintigraphy vs. 18F-Labeled NaF PET

TECHNICAL ASPECTS

Some commercial PET scanners allow data to be acquired in either 2-dimensional mode (interplane septa in place) or 3-dimensional mode (interplane septa removed). Because 18F-fluoride is well localized within the skeleton, excellent image quality can be obtained with either mode. The higher sensitivity of 3-dimensional PET allows the injected activity to be kept at a level (e.g., 2.1 MBq/kg) that yields a radiation dose to the patient similar to that received from 99mTc-MDP (Table 1). Moreover, the rapid uptake in bone and fast clearance from soft tissue of 18F-fluoride allows data recording to commence 15–30 min after intravenous administration of the radiopharmaceutical. Because modern PET scanners have axial fields of view of 15–20 cm, multiple bed positions will likely be required to adequately image the area of interest. If the area of concern is localized (e.g., an area of a suspected spinal injury), then 1 or 2 bed positions may be all that is required. In other cases (e.g., an oncologic survey scan), an extended study covering most or all of the skeleton may be required. Acquisition times of 3–5 min per bed position may be applied, depending on the sensitivity of the specific PET scanner being used, the administered activity, and whether the data are being acquired in 2-dimensional or 3-dimensional mode. Routine approaches to PET reconstruction such as filtered backprojection, or iterative approaches such as ordered-subset expectation maximization, can be effectively used with these data. In general, the same reconstruction parameters used for 18F-FDG studies also can be used for 18F-labeled NaF studies, although higher-frequency filtration also may be considered because of the high uptake of radioactivity and the fine spatial detail being imaged.

Because 18F-fluoride is well localized within the skeleton, excellent images can be reconstructed without the use of attenuation correction. The extent of the radioactivity in the transverse plane, whether in the spine or in long bones, is typically small, creating little variation in the intensity of the signal because of photon attenuation. However, failure to use attenuation correction sometimes may lead to artifacts in the reconstructed data (Fig. 9). Although the extent of the activity may be small in the transverse plane, the extent may be quite large in the axial direction. Thus, attenuation can vary significantly when either sagittal or coronal slices are being viewed. Further, without attenuation correction, the thoracic spine will appear to have substantially more uptake than the lumbar spine because the lungs produce less photon attenuation than does soft tissue. Urinary accumulation and excretion of 18F-fluoride can result in image artifacts. Tracer accumulation in the bladder could obscure lesions in the sacrum or pelvis. Thus, as with 99mTc-MDP bone scans, it is helpful to have the patient void just before imaging. Accumulation of 18F-fluoride in the renal pelvis can produce a signal that is more intense than the signal in adjacent spine. Without attenuation correction, this tracer accumulation may produce streak artifacts. These artifacts result when the signal is attenuated in the right–left direction substantially more than in the anterior–posterior direction, limiting the ability of the reconstruction algorithm to adequately compensate for these streak artifacts. This effect can cause a substantial reduction in the PET signal in the adjacent spine, even when an iterative reconstruction algorithm such as ordered-subset expectation maximization is used (29). The application of attenuation correction yields projection data that have more angular uniformity. Calculated or measured attenuation correction using either rotating rods or CT-based attenuation correction will alleviate these artifacts. Thus, the decision to use attenuation correction may depend on which regions of the body are being imaged.

FIGURE 9. 
FIGURE 9. 

In 18F-fluoride PET bone scan without attenuation correction (top row, with coronal, sagittal, and transverse slices appearing from left to right), streak artifact caused by activity in renal collecting system leads to apparent loss of signal in right lumbar spine (solid arrows). Apparent increased signal in thoracic spine results from reduction of attenuation in lung region (open arrow). Both artifacts are greatly reduced when attenuation correction is applied using rotating rod sources of 68Ge/68Ga (middle row). After patient voided, repeated image (bottom row) acquired in single bed position shows resolution of artifact caused by activity in renal collecting system.

In some applications, the quantitative capability of PET may add value to an image. The capability to report the absolute uptake of 18F-fluoride is the result of the more straightforward attenuation correction for PET and the ability to calibrate the PET scanner to measure absolute activity concentration (in Bq/cm3). By comparison, SPECT typically provides only relative quantitation of radiopharmaceutical uptake. The quantitative capability of 18F-fluoride PET may prove particularly useful if the clinical interpretation depends on comparison to known standard levels of metabolic activity or on quantitative comparison to a prior study.

PRACTICAL ISSUES

The imaging time required for 18F-fluoride PET is significantly shorter than that required for 99mTc-diphosphonate scintigraphy. The average time from injection to imaging with 18F-labeled NaF is 15–30 min, compared with 3–4 h for 99mTc-MDP. Additionally, the total imaging time for 18F-fluoride PET is 15–30 min, whereas a 99mTc-MDP SPECT scan takes an average of 60 min. In general, the total examination time is approximately 1 h for 18F-fluoride PET, compared with approximately 4 h for 99mTc-MDP bone scintigraphy. Moreover, the reduced imaging time associated with 18F-fluoride PET minimizes the opportunity for patient motion and resulting imaging artifacts. However, if motion artifacts do occur, software correction cannot be performed with PET images, as sometimes is possible with SPECT. The use of PET may raise the question of sedation in young children to reduce patient motion. In our department, with skilled nuclear medicine technologists and the assistance of a trained child life specialist, sedation is rarely necessary for 99mTc-MDP SPECT bone scans. With shorter scan times, sedation is even less likely to be needed with 18F-fluoride PET.

Both the patient and the referring physician benefit from shorter examination times, which are more convenient, pose less physical and emotional stress, and accelerate diagnosis and treatment. With the more rapid turnaround of 18F-fluoride PET results, the patient can easily return the same day to discuss results with the referring physician. The shorter imaging time for 18F-fluoride PET also improves workflow efficiency and resource use. For example, 18F-fluoride PET can be scheduled in any available 1-h opening in the PET schedule, freeing an additional hour of γ-camera imaging time for other scintigraphic studies. At present, the cost per dose of 18F-labeled NaF is higher than that of 99mTc-MDP, but this difference is at least partially offset by the fewer resources needed for shorter imaging times. In addition, as with 18F-FDG, the unit dose cost of 18F-labeled NaF is likely to decrease as demand increases and the overall market for 18F-labeled NaF grows.

At present, 18F-fluoride PET typically is not performed outside large medical centers because of widespread uncertainty about insurance reimbursement. Although U.S. Food and Drug Administration approval and accepted clinical use of 18F-labeled NaF for skeletal imaging predates those of 99mTc diphosphonates, many insurance guidelines lack a description of 18F-fluoride bone imaging with any camera type, and most insurance coverage for tomographic imaging is limited to 99mTc-diphosphonate SPECT. Further contributing to these obstacles is the lack of a Current Procedural Terminology code for 18F-fluoride PET for bone imaging. Assignment of such a Current Procedural Terminology code should facilitate efforts to encourage individual insurance companies to develop specific guidelines for reimbursement.

CONCLUSION

18F-labeled NaF, although one of the early bone-scanning agents, was displaced by the arrival of 99mTc-labeled bone-imaging agents, which were available in convenient kit form for labeling with generator-produced 99mTc and produced better images on the commonly used, Anger-style, γ-cameras. The widespread availability of modern PET scanners permits high-quality skeletal imaging with 18F-fluoride, which has the favorable characteristics of highly specific bone uptake, rapid clearance from the blood pool because of minimal protein binding, and dosimetry similar to that of 99mTc-MDP. The commercial delivery system used for 18F-FDG also can be used for the efficient delivery of other 18F-labeled radiopharmaceuticals, such as 18F-labeled NaF. It is now feasible to perform high-quality 18F-labeled NaF bone scans in most nuclear medicine departments. Numerous recent studies have compared 18F-fluoride PET to 99mTc-MDP scintigraphy. These studies have demonstrated that 18F-fluoride PET is more accurate than planar imaging or SPECT with 99mTc-MDP for localizing and characterizing both malignant and benign bone lesions. The addition of correlative imaging, such as CT, MRI, or hybrid imaging with PET/CT, further improves the specificity and accuracy of 18F-fluoride skeletal PET. Direct correlation of 18F-fluoride PET and anatomic imaging using either fusion software or hybrid imaging probably will become the routine clinical practice in nearly all cases. The clinical usefulness of 18F-fluoride PET has been demonstrated for a wide range of clinical indications for oncology and for benign diseases of bone (Table 2). Although not yet in routine clinical use, quantitative 18F-fluoride PET may prove useful for the assessment of metabolic bone disorders such as renal osteodystrophy, osteoporosis, or Paget's disease. For many clinical indications, 18F-fluoride PET images can be obtained without attenuation correction, although some well-defined artifacts may be avoided by applying attenuation correction.

View this table:
TABLE 2

Indications for 18F-Fluoride Skeletal PET

18F-Fluoride PET offers the additional advantages of faster study times, improved workflow in the nuclear medicine clinic, increased convenience to the patient, and rapid turnaround of results to the referring physician. The widespread readoption of 18F-labeled NaF as a bone-imaging agent, however, has been limited by longstanding familiarity with 99mTc-diphosphonate scintigraphy and by issues related to insurance reimbursement for 18F-fluoride PET. The higher-quality imaging, increased clinical accuracy, greater convenience to the patient and referring physician, and more efficient use of nuclear medicine resources all indicate the need to reconsider the use of 18F-fluoride PET for imaging benign and malignant diseases of the skeleton.

Acknowledgments

We acknowledge the technical support of Stephen Laffin and of the rest of the technical team in the Division of Nuclear Medicine at Children's Hospital Boston. We thank Nancy Drinan and Christopher Durall for editorial assistance in the preparation of the manuscript. Edmond Reidy, Radiology Administrator, helped with the economic assessment.

Footnotes

  • ↵* NOTE: FOR CE CREDIT, YOU CAN ACCESS THIS ACTIVITY THROUGH THE SNM WEB SITE (http://www.snm.org/ce_online) THROUGH JANUARY 2009.

  • No potential conflict of interest relevant to this article was reported.

  • COPYRIGHT © 2008 by the Society of Nuclear Medicine, Inc.

References

  1. 1.↵
    Blau M, Ganatra R, Bender MA. 18 F-Fluoride for bone imaging. Semin Nucl Med. 1972;2:31–37.
    OpenUrlCrossRefPubMed
  2. 2.↵
    Blau M, Nagler W, Bender MA. Fluorine-18: a new isotope for bone scanning. J Nucl Med. 1962;3:332–334.
    OpenUrlPubMed
  3. 3.↵
    Thrall JH. Technetium-99m labeled agents for skeletal imaging. CRC Crit Rev Clin Radiol Nucl Med. 1976;8:1–31.
    OpenUrlPubMed
  4. 4.↵
    Davis MA, Jones AG. Comparison of 99mTc-labeled phosphate and phosphonate agents for skeletal imaging. Semin Nucl Med. 1976;6:19–31.
    OpenUrlPubMed
  5. 5.↵
    Hawkins RA, Choi Y, Huang SC, et al. Evaluation of the skeletal kinetics of fluorine-18-fluoride ion with PET. J Nucl Med. 1992;33:633–642.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    Hoh CK, Hawkins RA, Dahlbom M, et al. Whole body skeletal imaging with [18F]fluoride ion and PET. J Comput Assist Tomogr. 1993;17:34–41.
    OpenUrlCrossRefPubMed
  7. 7.↵
    Anbar M, Ernst N. A distribution study of F18-labelled cationic fluorocomplexes in rats. Int J Appl Radiat Isot. 1962;13:47–51.
    OpenUrlCrossRef
  8. 8.↵
    Ruth TJ, Wolf AP. Absolute cross sections for the production of fluorine-18 via the 18O(p,n)18F reaction. Radiochimica Acta. 1979;26:21–24.
    OpenUrl
  9. 9.↵
    Weber DA, Greenberg EJ, Dimich A, et al. Kinetics of radionuclides used for bone studies. J Nucl Med. 1969;10:8–17.
    OpenUrlFREE Full Text
  10. 10.↵
    Wootton R, Dore C. The single-passage extraction of 18F in rabbit bone. Clin Phys Physiol Meas. 1986;7:333–343.
    OpenUrlCrossRefPubMed
  11. 11.↵
    Costeas A, Woodard HQ, Laughlin JS. Depletion of 18F from blood flowing through bone. J Nucl Med. 1970;11:43–45.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    Blake GM, Park-Holohan SJ, Cook GJ, Fogelman I. Quantitative studies of bone with the use of 18F-fluoride and 99mTc-methylene diphosphonate. Semin Nucl Med. 2001;31:28–49.
    OpenUrlCrossRefPubMed
  13. 13.↵
    Chilton HM, Francis MD, Thrall JH. Radiopharmaceuticals for bone and bone marrow imaging. In: Swanson DP, Chilton HM, Thrall JH, eds. Pharmaceuticals in Medical Imaging: Radiopaque Contrast Media, Radiopharmaceuticals, Enhancement Agents for Magnetic Resonance Imaging and Ultrasound. New York, NY: Macmillan Pub. Co.; 1990:537–563.
  14. 14.↵
    Hyldstrup L, McNair P, Ring P, Henriksen O. Studies on diphosphonate kinetics. Part I: evaluation of plasma elimination curves during 24 h. Eur J Nucl Med. 1987;12:581–584.
    OpenUrlPubMed
  15. 15.↵
    Park-Holohan SJ, Blake GM, Fogelman I. Quantitative studies of bone using 18F-fluoride and 99mTc-methylene diphosphonate: evaluation of renal and whole-blood kinetics. Nucl Med Commun. 2001;22:1037–1044.
    OpenUrlCrossRefPubMed
  16. 16.↵
    Tse N, Hoh C, Hawkins R, Phelps M, Glaspy J. Positron emission tomography diagnosis of pulmonary metastases in osteogenic sarcoma. Am J Clin Oncol. 1994;17:22–25.
    OpenUrlPubMed
  17. 17.↵
    Petren-Mallmin M, Andreasson I, Ljunggren O, et al. Skeletal metastases from breast cancer: uptake of 18F-fluoride measured with positron emission tomography in correlation with CT. Skeletal Radiol. 1998;27:72–76.
    OpenUrlCrossRefPubMed
  18. 18.↵
    Schirrmeister H, Guhlmann A, Elsner K, et al. Sensitivity in detecting osseous lesions depends on anatomic localization: planar bone scintigraphy versus 18F PET. J Nucl Med. 1999;40:1623–1629.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    Schirrmeister H, Guhlmann A, Kotzerke J, et al. Early detection and accurate description of extent of metastatic bone disease in breast cancer with fluoride ion and positron emission tomography. J Clin Oncol. 1999;17:2381–2389.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    Schirrmeister H, Glatting G, Hetzel J, et al. Prospective evaluation of the clinical value of planar bone scans, SPECT, and 18F-labeled NaF PET in newly diagnosed lung cancer. J Nucl Med. 2001;42:1800–1804.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    Hetzel M, Arslandemir C, Konig HH, et al. F-18 NaF PET for detection of bone metastases in lung cancer: accuracy, cost-effectiveness, and impact on patient management. J Bone Miner Res. 2003;18:2206–2214.
    OpenUrlCrossRefPubMed
  22. 22.↵
    Hoegerle S, Juengling F, Otte A, Altehoefer C, Moser EA, Nitzsche EU. Combined FDG and [F-18]fluoride whole-body PET: a feasible two-in-one approach to cancer imaging? Radiology. 1998;209:253–258.
    OpenUrlPubMed
  23. 23.↵
    Langsteger W, Heinisch M, Fogelman I. The role of fluorodeoxyglucose, 18F-dihydroxyphenylalanine, 18F-choline, and 18F-fluoride in bone imaging with emphasis on prostate and breast. Semin Nucl Med. 2006;36:73–92.
    OpenUrlCrossRefPubMed
  24. 24.↵
    Even-Sapir E, Metser U, Flusser G, et al. Assessment of malignant skeletal disease: initial experience with 18F-fluoride PET/CT and comparison between 18F-fluoride PET and 18F-fluoride PET/CT. J Nucl Med. 2004;45:272–278.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    Even-Sapir E, Metser U, Mishani E, Lievshitz G, Lerman H, Leibovitch I. The detection of bone metastases in patients with high-risk prostate cancer: 99mTc-MDP planar bone scintigraphy, single- and multi-field-of-view SPECT, 18F-fluoride PET, and 18F-fluoride PET/CT. J Nucl Med. 2006;47:287–297.
    OpenUrlAbstract/FREE Full Text
  26. 26.↵
    Wade AA, Scott JA, Kuter I, Fischman AJ. Flare response in 18F-fluoride ion PET bone scanning. AJR. 2006;186:1783–1786.
    OpenUrlCrossRefPubMed
  27. 27.↵
    Coady CM, Micheli LJ. Stress fractures in the pediatric athlete. Clin Sports Med. 1997;16:225–238.
    OpenUrlCrossRefPubMed
  28. 28.↵
    Ovadia D, Metser U, Lievshitz G, Yaniv M, Wientroub S, Even-Sapir E. Back pain in adolescents: assessment with integrated 18F-fluoride positron-emission tomography-computed tomography. J Pediatr Orthop. 2007;27:90–93.
    OpenUrlPubMed
  29. 29.↵
    Lim R, Fahey FH, Drubach LA, Connolly LP, Treves ST. Early experience with 18F sodium fluoride bone PET in young patients with back pain. J Pediatr Orthop. 2007;27:277–282.
    OpenUrlPubMed
  30. 30.↵
    Schiepers C, Broos P, Miserez M, Bormans G, De Roo M. Measurement of skeletal flow with positron emission tomography and 18F-fluoride in femoral head osteonecrosis. Arch Orthop Trauma Surg. 1998;118:131–135.
    OpenUrlPubMed
  31. 31.↵
    Forrest N, Welch A, Murray AD, Schweiger L, Hutchison J, Ashcroft GP. Femoral head viability after Birmingham resurfacing hip arthroplasty: assessment with use of [18F] fluoride positron emission tomography. J Bone Joint Surg Am. 2006;88(suppl 3):84–89.
    OpenUrlCrossRefPubMed
  32. 32.↵
    Piert M, Winter E, Becker GA, et al. Allogenic bone graft viability after hip revision arthroplasty assessed by dynamic [18F]fluoride ion positron emission tomography. Eur J Nucl Med. 1999;26:615–624.
    OpenUrlCrossRefPubMed
  33. 33.↵
    Sorensen J, Ullmark G, Langstrom B, Nilsson O. Rapid bone and blood flow formation in impacted morselized allografts: positron emission tomography (PET) studies on allografts in 5 femoral component revisions of total hip arthroplasty. Acta Orthop Scand. 2003;74:633–643.
    OpenUrlCrossRefPubMed
  34. 34.↵
    Messa C, Goodman WG, Hoh CK, et al. Bone metabolic activity measured with positron emission tomography and [18F]fluoride ion in renal osteodystrophy: correlation with bone histomorphometry. J Clin Endocrinol Metab. 1993;77:949–955.
    OpenUrlCrossRefPubMed
  35. 35.↵
    Piert M, Zittel TT, Becker GA, et al. Assessment of porcine bone metabolism by dynamic 18F-fluoride ion PET: correlation with bone histomorphometry. J Nucl Med. 2001;42:1091–1100.
    OpenUrlAbstract/FREE Full Text
  36. 36.↵
    Blake GM, Park-Holohan SJ, Fogelman I. Quantitative studies of bone in postmenopausal women using 18F-fluoride and 99mTc-methylene diphosphonate. J Nucl Med. 2002;43:338–345.
    OpenUrlAbstract/FREE Full Text
  37. 37.↵
    Installe J, Nzeusseu A, Bol A, Depresseux G, Devogelaer JP, Lonneux M. 18F-Fluoride PET for monitoring therapeutic response in Paget's disease of bone. J Nucl Med. 2005;46:1650–1658.
    OpenUrlAbstract/FREE Full Text
  38. 38.↵
    ICRP Publication 53: Radiation Dose to Patients from Radiopharmaceuticals. Stockholm, Sweden: International Commission on Radiological Protection; 1987.
  39. 39.↵
    ICRP Publication 80: Radiation Dose to Patients from Radiopharmaceuticals. Stockholm, Sweden: International Commission on Radiological Protection; 1999.
  • Received for publication May 4, 2007.
  • Accepted for publication September 10, 2007.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles