18F NaF PET/CT andConventional Bone Scanningin Routine Clinical PracticeComparative Analysis of Tracers, ClinicalAcquisition Protocols, and PerformanceIndices

Gad Abikhzer, MDCM, FRCPC, ABNMa,*,1,John Kennedy, PhDa, Ora Israel, MDa,b

KEYWORDS

� Bone scan � F-18 fluoride � PET/CT � Whole body SPECT

KEY POINTS

� Conventional planar and SPECT bone scintigraphy is an established technique.

� F-18 fluoride PET/CT has the potential to become the gold standard in functional bone imaging.

� A thorough comparison of both techniques, including advantages and disadvantages arediscussed.

� Future directions of both modalities are analyzed.

INTRODUCTION

Bone scintigraphy (BS) was one of the ear-liest examinations performed in nuclear medicine.F-18 sodium fluoride (NaF) is a bone-scanningagent that was first introduced in 1962.1 The high511 keV energy annihilation photons emitted byF-18 could be imaged at that time with rectilinearscanners equipped with thick crystals. For presentday standards, the images obtained were of poorquality. With the advent of the first technetium-99m (Tc)-based phosphonates in 1971 followedby methylene disphosphonate (MDP)2 and thedevelopment of the Anger camera, fluoride bonescanning was replaced. BS has become one ofthe most common procedures, widely used in the

The authors have nothing to disclose.a DepartmentofNuclearMedicine, RambamHealth Care CaRapaport Faculty of Medicine, Technion–Israel Institute of1 Dr Abikhzer is a clinical and research fellow supportscholarship fund of the Rambam Health Care Campus R&* Corresponding author.E-mail address: gad101@hotmail.com

PET Clin 7 (2012) 315–328doi:10.1016/j.cpet.2012.04.0051556-8598/12/$ – see front matter � 2012 Elsevier Inc. All

evaluation of malignant and benign diseases ofthe skeleton. The advent of single-photon emissiontomography (SPECT) and eventually SPECT/com-puted tomography (CT) has further increased thediagnostic accuracy of BS and its clinical applica-tions.3 Over the past few years, with the rapidlyincreasing implementation of PET/CT devices andF-18, there has been a reemergence of interestand use of NaF.

There are now 2 excellent bone-scanning agentsavailable, and the nuclear medicine community isfaced with the dilemma of which one to adopt inroutine clinical use. Tc-MDP has withstood thetest of time, is easily available from generatorseven in remote locations, and can be used withtheeasily accessible gammacameras that currently

mpus, 6 Ha’Aliya Street, Haifa 31096, Israel; b B. and R.Technology, Haifa, Israeled by the Azrieli foundation and the Dr Dov FrontD foundation.

rights reserved. pet.theclinics.com

Abikhzer et al316

outnumber PET/CT devices. On the other hand,PET/CT devices are becoming readily availableworldwide, together with cyclotrons and distribu-tion networks for 18F-fluorodeoxyglucose (FDG),and this has facilitated as well the use of NaF inroutine clinical work. Although there are a numberof advantages to the use of NaF, which will bedescribed in this article, do theexcellent pharmaco-kinetic properties of this tracer and superior resolu-tionofPET/CTdevices translate toclinical benefits?

PHARMACOKINETICS AND UPTAKEMECHANISM

Comparison of the pharmacokinetics of Tc-MDPand NaF offer a theoretical advantage for NaF.First-pass clearance of Tc-based phosphonates isapproximately 64%.4 Protein binding of Tc-MDP is25% to 30% immediately after injection andapprox-imately50%at4hours.5 It doesnotbind to redbloodcells (RBCs). ImagingwithTc-MDPrequiresa2-hourto 4-hour uptake time after injection, when 40% ofthe tracer is found in skeleton, 40% in urine, 10%in soft tissues, and only 5% in the blood stream,which results in improved target-to-backgroundratio and better image quality.2 NaF undergoesmore rapid blood clearance, with first-pass clear-ance close to 100%,6 as it has negligible proteinbinding. Approximately 30% of the injected dose isin RBCs, which does not interfere with bone uptake,as NaF freely diffuses across the cell membrane.7

Plasma clearance is very rapid. Approximately50% of the injected NaF is taken up in bone,8 withthe remainder excreted by the kidneys by 6 hoursafter tracer administration.9 These properties permita shorter uptake time of NaF with earlier start ofimaging (Table 1).

Bone is composedof two-thirdsmineral andone-third collagen, extracellular matrix, and a variety of

Table 1Comparison of pharmacokinetic properties ofTc-MDP and NaF

MDP NaF

Urinaryexcretion

70% after 6 h4 50% after 6 h9

Proteinbinding

50% at 4 h5 Negligible7

RBC binding Negligible5 30%7

First-passclearance

w64%4 Nearly 100%6

% boneuptake

35%–50%2 50%8

Abbreviations: MDP, methylene disphosphonate; NaF,sodium fluoride; Tc, technetium.

bone-lining cells. The mineral matrix is composedof calcium hydroxyapatite, Ca10(OH)2(PO4)6, con-taining calcium phosphate that can be exchangedwith phosphonates present in MDP.2

Tc-MDP uptake in bone is considered to bemainly related to chemisorption of the disphospho-nate onto the surface of hydroxyapatite, followedby incorporation into the crystal. Bone uptake ofTc-MDP is related to increased blood flow andcapillary permeability, as well as to increasedbone turnover with osteoid formation. Uptake ofTc-MDP in immature collagen has also beendescribed.2UptakeofNaFhasasimilarmechanismto Tc-MDP. Following chemisorption of fluorideions onto the surface of hydroxyapatite, theyexchangewith the hydroxyl (OH–) ions in the crystal,forming fluoroapatite.4

Imaging Protocols

Requisition and historyRequisition for BS and NaF PET/CT by the referringphysician should ideally be accompanied by a con-cise summaryof thepatient’s historywith apertinentclinical question. The following points are very im-portant for accurate interpretation of the examina-tion, depending on the clinical indication.

� History of malignancy� Date of recent chemotherapy� Previous fractures or recent trauma� Previous orthopedic surgery and relevant

dates� Previous infection and its location� Urinary diversion procedures� Location of any bone pain.

Previous bone scintigraphy, as well as otherrelevant imaging studies, should be available forcomparison.10

Patient preparationThe preparation of patients referred for BS andNaFPET/CT is essentially the same. Patients should bewell hydrated before the study and during theuptake period between the time of tracer injectionand imaging. This enhances renal excretion, result-ing in improved target-to-background ratio andreducing radiation exposure of the patients.Patients are also encouraged to drink morefrequently for the remainder of the day. Immedi-ately before image acquisition, patients are askedto empty their bladders. Any metal objects shouldbe removed to prevent attenuation artifacts.10

Radiopharmaceutical, injected activity, anduptake timeBS using Tc-MDP (other phosphonate-basedtracers are also available, such as di-carboxy

Bone Scanning in Routine Clinical Practice 317

propane-diphosphonate [DPD] and hydroxy-methylene disphosphonate [HDP]):

� Adults: 740 to 1110 MBq (20–30 mCi) is in-jected intravenously (IV). The administeredactivity shouldbe increased inobesepatientsto 11 to 13 MBq/kg (300–350 mCi/kg). Injec-tion should be performed ideally with an IVcatheter rather than through direct venipunc-ture, as injection site infiltrations can causediscomfort to the patient and result in recon-struction streak artifacts, most notably withSPECT.

� Pediatrics: 9 to 11MBq/kg (250–300 mCi/kg)is administeredwith aminimumactivity of 20to 40 MBq (0.5–1.0 mCi) and maximumactivity that should not exceed adultdoses.10

Tc-MDP preparation uses simple, commercialkits, thus permitting greater department flexibilitywhen additional studies are scheduled at the lastminute or emergency BS is requested (Table 2).

PET/CT using NaF:

� Adults: 185 to 370 MBq (5–10 mCi) is in-jected IV, through a catheter. For obesepatients, injected activity should be at theupper limit of recommended doses.

� Pediatrics: 2.59 MBq/kg, 70 mCi/kg isadministered with a minimum activity of 11MBq (0.3 mCi) and the maximum level notexceeding 185 MBq (5 mCi).11

The shorter half-life of NaF requires more rigidscheduling and lowers the ability to makeunexpected changes in the clinical daily schedule(see Table 2).

Oral administration of NaF has been used previ-ously. It appears to be as effective as IV adminis-tration, as the intestinal absorption of NaF isprompt and complete. Residual NaF in the gastro-intestinal tract may occasionally remain in certain

Table 2Comparison of administered dose, uptake time of raof Tc-MDP and NaF

Tc-99m MDP BS1

Dosage (adult) 740–1110 MBq (20–30 mCi)

Dosage (pediatric) 9–11 MBq/kg (250–300 mCi/kg)

Time to imaging 2–4 h

Radiation dosimetry 4.2 mSv (planar BS)7.4 mSv (SPECT/CT)

Abbreviations: BS, bone scintigraphy; CT, computed tomographSPECT, single-photon emission computed tomography; Tc-99m

patients with incomplete absorption.12 With cur-rent PET/CT technology, this should not causea diagnostic dilemma. This route of administrationmay represent an alternative to IV injection inpatients with difficult venous access.

Although peak bone activity for Tc-MDP is 1 hourpost injection, soft tissue clearance is nearlycomplete at 4 hours, with imaging generally per-formed at 2 to 4 hours post injection. NaF PET/CTis a more rapid overall examination. The uptaketime is usually 30 to 45 minutes for patients withnormal renal function. A 90-minute to 120-minutewait is recommended when imaging the lowerextremities.11

Image acquisitionBS and SPECT BS acquisition protocols vary de-pending on the routine work flows in differentdepartments. Standard planar scintigraphy is per-formed at 2 to 4 hours after injection. A 3-phase, oroccasionally a 4-phase study, can be also per-formed. The flow (first phase) and blood pool(second phase) studies add important informationto the standard BS (third phase). They assess thepresence of blood flow abnormalities (increasedperfusion is the more frequent finding) that mayoccur in various clinical settings, such as osteomy-elitis, complex regional pain syndrome, andheterotopic ossifications, as well as in primarybone tumors, avascular necrosis, Paget disease,or fractures. The 24-hour delayed phase (fourthphase) improves bone-to–soft tissue differentia-tion and is therefore useful in the elderly, inpatients with diabetes, or those with peripheralvenous disease in whom increased transit anduptake time impair image quality, mainly of thelower extremities.2 It can also be used when highamounts of the radiopharmaceutical retained inthe urinary tract (eg, a full urinary bladder) obscurepelvic bone structures, or when urinary contami-nation is present on the 3-hour scan and is

diopharmaceuticals, and radiation dosimetry

18F NaF PET/CT 11

185–370 MBq (5–10 mCi)

2.59 MBq/kg (70 mCi/kg)

30–45 minutes (90–120 minutes for extremities)

12.1 mSv

y; MDP, methylene disphosphonate; NaF, sodium fluoride;, technetium-99m.

Abikhzer et al318

expected to disappear on delayed images. Withthe use of SPECT imaging, this phase is less oftenrequired.SPECT imaging as an adjunct to BS improves

contrast resolution, and enables better separationof adjacent structures and identification of lesionsobscured by the 2-dimensional planar images.Sensitivity of BS has improved with SPECT, with20% to 50% more vertebral lesions, for example,being detected than with planar BS only. Speci-ficity of BS with SPECT is also higher because ofmore accurate lesion localization.3

Low-energy, high-resolution (LEHR) collimatorsare used for whole-body planar BS and SPECT.Flow/perfusion and blood pool scans aiming atdetection of hyperemia are performed immediatelyafter the injection of Tc-MDP, including a less than1-minute dynamic study (eg, 16 framesat 2 secondsper frame), immediately followed by approximately300 kcounts anterior and posterior planar scans(1024 � 256 matrix) of a specific region of interest.The planar whole-body scan is performed usingcontinuous bed motion for 1500 kcounts with a 256� 1024 matrix. A typical SPECT bone-acquisitionprotocol uses 2 detectors for 60 to 120 stops overa 360� scan with 10 to 40 seconds per stop. Recon-struction matrix is at least 64 � 64 (Table 3).10

With conventional gamma cameras, routineacquisition time for planar BS of the entire skeletonis approximately 12 minutes, followed by a singlefield-of-view (FOV) SPECT of about 15 minutes.SPECT acquisition time can be reduced by a factorof 2 while maintaining image quality13 with the useof collimator-detector response compensation inimage reconstruction, also known as resolutionrecovery. Iterative reconstruction is typically usedand images are post-filtered to reduce noise. Avail-able reconstruction software products includeWide Beam Reconstruction (UltraSPECT, Haifa,

Table 3Sample image acquisition protocol for BS

Collimator LEHR

Flow 2 s/frame for a total of 32seconds

Blood pool w300 kcount anterior andposterior scans

Planar skeletalphase

1500 kcounts whole-body scan

SPECT 60–120 stops, over a 360� scan,10–40 s per stop, 64 � 64matrix reconstruction

Abbreviations: BS, bone scintigraphy; LEHR, low-energy,high-resolution; SPECT, single-photon emission computedtomography.

Israel), Evolution (GE Healthcare, Waukesha,WI, USA), Astonish (Philips Healthcare, Best,Netherlands), and Flash 3D (Siemens Healthcare,Erlangen, Germany).

NaF PET To obtain high-quality skeletal imageson current devices, emission scans withacquisition times of 1 to 2 minutes per bed positionin 3-dimensional mode are obtained, dependingon the camera, injected activity, uptake time, andbody habitus of the patient. For a scan rangefrom the skull to upper thighs with an average of9 bed positions, acquisition time is in the rangeof 9 to 18 minutes, including CT acquisition andbed translation, longer for a whole body study,which requires an average of 15 bed positions.Reconstruction typically uses 3-dimensional maxi-mum likelihood expectation maximization (MLEM)algorithm and a 128 � 128 or 256 � 256 matrix,with post filtering. A maximum intensity projection(MIP) image that is generated can facilitate detec-tion of suspicious regions of interest for diagnosis(Fig. 1, Table 4).11

Dynamic studies for NaF PET studies have beenperformed for investigational purposes to pro-vide quantitation of kinetic NaF incorporationparameters.14,15

Hybrid imaging of the bone Hybrid scannersprovide the ability to perform emission and CTscans sequentially in a single imaging session ona single imaging device. For PET/CT, whole-bodyhelical CT is acquired either immediately before orafter the emission scan to provide CT-based atten-uation correction and anatomic localization.Typical settings for an attenuation correction/local-izationCT scan are a tube current of 30mA, voltageof 120 kVp, rotation of 0.5 second, and pitch of 1.11

CT parameters can be modified, aiming at eitherreduced radiation exposure to the patient, associ-ated with a poorer image quality, or alternatively,to provide diagnostic studies if required.SPECT/CT is typically limited toa regionof interest

defined by suspicious findings in the skeleton de-tected on the planar component of the study, orguided by specific patient symptoms or previousfindings. The duration of theCTacquisition dependson the type of CT component of the specific imagingdevice, ranging from 5minutes for 1 bed position fora hybrid SPECT/CT with an x-ray tube mounted onthe same gantry as the gamma camera detectors,to less than 1 minute for a hybrid PET or SPECTdevice with a multislice CT.

Radiation Dosimetry

The effective dose to the patient from an injectionof 370 MBq of NaF is about 8.9 mSv compared

Fig. 1. Normal 18F NaF PET/CT with coronal (A), axial (B), sagittal (C), and MIP (D) reconstructions. Note slightlyheterogeneous bone tracer distribution throughout the skeleton.

Bone Scanning in Routine Clinical Practice 319

with 4.2 mSv for 740MBq of Tc-MDP for SPECT.11

A meta-analysis reported a range of values of 2.7to 15.0 mSv for NaF versus 4.2 to 5.7 mSv forTc-MDP, respectively.16 In children, these valuesare lower.17 The radiation exposure associatedwith the CT component of PET/CT and SPECT/CT studies is highly variable and ranges fromless than 1 mSv for CT-attenuation correction

only,18 to about 8 mSv for a diagnostic CTscan.19 A typical value for an effective CT doseused for localization and attenuation correction is3.2 mSv.11 Consequently, the total effectivedose of a NaF PET with a localization CT bonescan is approximately 12.1 mSv compared withapproximately 7.4 mSv for a bone SPECT/CT(see Table 2).

Table 4Sample image acquisition protocol with NaF PET

No. of bed positions 9–15

Time/bed position 1–2 minutes

Acquisition mode 3 dimensional

Reconstruction MLEM, 128 � 128 or256 � 256 matrix

Abbreviations: MLEM, maximum likelihood expectationmaximization; NaF, sodium fluoride.

Abikhzer et al320

Technological Advances and Novel Protocols

Spatial resolution of clinical images is dependenton camera design, image reconstruction, andpostprocessing. The current intrinsic spatial reso-lution of PET is considerably better, about 5 mm,as compared with approximately 10 mm forSPECT.20 Although the lower resolution of SPECTdecreases the ability to detect small lesionsbecause of partial volume effects,21 these sitesmay still be identified if the radiotracer uptake isintense relative to the surrounding tissue.The recent software developments discussed

previously currently allow routine acquisition ofwhole body (WB) SPECT, which permits a tomo-graphic assessment of the entire skeleton (Fig. 2),similar to the images obtained with 18F NaF PET/CT(see Fig. 1). Tomographic WB SPECT can be ob-tained with Tc-MDP, with acquisition times that aresomewhat longer but within a comparable range toNaFPET/CT.Ascan range fromtheskull to thekneesthat requires 3 bed positions, spanning about 1.2 maxially, can be performedwithin a significantly short-er acquisition time (eg, 7.5 minutes per bed positionfor a total of 22.5 minutes), and will probably de-crease even more in the future.Thedevelopment ofPETscannerswith additional

detector rings increases the axial FOV and reducesimageacquisition times.Commercial scannerswithextended axial FOVs canperforma skeletal study in7.5 minutes, and future newly designed deviceswould use 1 bed position to scan the entire patientlength in less than 6 minutes.22 Routine CT acquisi-tion obtained with PET/CT increases the studyspecificity and represents an additional advantageof NaF PET/CT compared with BS.23

WB PET/magnetic resonance (MR) imagingscanners have also been developed. WB MRimaging sequences are very accurate for the eval-uation of bone and bone marrow metastases.24 Ahybrid PET/MR scan following the injection ofNaF and/or FDG may represent in the future the1-stop bone-imaging modality.Sequential injection and imaging of FDGandNaF

has been described, attempting to increase the

diagnostic accuracy for bone metastases. Thisprotocol involves the near simultaneous injectionof 555 MBq (15 mCi) of FDG and 185 MBq (5 mCi)of NaF. PET/CT images are acquired followinga 60-minute uptake time. A bone mask is derivedfrom the CT data and multiplied with the PET datato produce an image of the skeleton witha combined NaF/FDG signal, excluding soft tissueuptake. A pilot study demonstrated the feasibilityof combined NaF/FDG-PET/CT imaging showingsimilar accuracy for the detection of bonemetastases compared with separate interpretationof each modality.25 The investigators advocate theuse of the combined technique to increasepatient convenience and reduce radiation dose(as only 1 CT is performed). In a larger patient pop-ulation, the same group found that the combinedNaFandFDGPET/CTscan increased thesensitivityfor detection of bone metastases when comparedwith FDG PET/CT only.26 It has been suggestedthat this technique may obscure bone marrowmetastases.11 Although it remains tobedeterminedwhether this protocol will be clinically useful,27 evenif validated it does not necessarily represent anadvantage of NaF PET/CT over conventional BS.A 1-day protocol combining conventional BS andFDG PET/CT is also technically feasible,28,29

despite increased photon scatter and detectordead time,because the2 typesofdetectedphotonshave significantly different energies (511 keV and140 keV respectively).

Normal Biodistribution and ImageInterpretation

Interpretation and normal variants of conventionalBS are well known to the nuclear medicine physi-cian and will not be further discussed in this article(see Fig. 2). An excellent review discussescommon patterns, variants, and artifacts with BSand SPECT/CT.30

A normal NaF PET demonstrates slight nonho-mogeneity in tracer distribution throughout theskeleton (see Fig. 1). This reflects differences inregional blood flow, as well as differences in thebone crystal surface area accessible to thetracer.12 Urinary excretion of the tracer results inkidney and urinary bladder visualization.Interpretation of both types of bone scans is

similar, with the same patterns of abnormal uptaketo be expected (Figs. 3 and 4). Any causes ofaltered bone metabolism may cause increasedNaF uptake (Fig. 5), which at present is most oftenused for metastatic disease (see Fig. 3). Uptakemay be quite prominent in benign findings aswell. NaF PET/CT has a high sensitivity for detect-ing lytic lesions (Fig. 6).31 For those who are

Fig. 2. Normal Tc-MDP BS in the same patient as in Fig. 1: Planar (A) and WB SPECT with coronal (B), axial (C),sagittal (D), and MIP (E) reconstructions.

Bone Scanning in Routine Clinical Practice 321

routinely reading BS, a learning curve is associ-ated with the interpretation of NaF PET/CT beforebecoming familiar with the normal and sometimesprominent heterogeneity of tracer distribution.

Only a few studies have assessed NaF standarduptake values (SUVs), which are not routinely used

in the interpretation of NaF PET/CT at present.One study measured the SUVmax of NaF uptakein metastatic lesions in a variety of malignancies.Sclerotic or mixed lesions had higher SUV valuesthan lytic sites. The same was found for lesionsinvolving both the cortex and medulla when

Fig. 3. Patient with extensive metastatic breast cancer imaged with NaF PET/CT. Sagittal (A), axial (B), and MIP (C)images are shown.

Abikhzer et al322

compared with those involving only the cortex orthe medulla.32 In a study comparing NaF PET/CTto F-18 fluorocholine (FCH) in patients with pros-tate cancer, SUVmax was measured in all bonemetastases. With NaF, mean SUVmax in malig-nant lesions was 57� 43 as compared with 20� 7for benign lesions. Although SUVmax was not ableto differentiate between benign and malignantlesions, none of the benign lesions had an SUV-max higher than 45.33 The applicability of thisstatement in routine clinical work is questionable,as certain benign bone entities, including fibrousdysplasia, Paget disease, and acute fractures,can demonstrate very intense uptake. Intervalchanges in SUVmax have been found in a fewpatients to be able to predict response to therapy,even in the absence of differences on visual

comparison.34 It has not yet been possible toaccurately quantify SPECT uptake of Tc-MDP inroutine clinical work.35

Clinical Applications and Performance of BSand NaF PET/CT

This issue of the PET Clinics reviews the clinicalutility of NaF PET in the assessment of a varietyof bone diseases. Various clinical indications andpotential applications for NaF PET/CT and/or BSare summarized in Box 1 and are reviewed indepth in the appropriate articles.To accurately assess both technologies,

studies have to compare similar state-of-the-arttechnologies. This is analogous to the classic idiomof the need to compare apples with apples

Fig. 4. Patient with extensive metastatic breast cancer imaged with Tc-MDP BS (same patient as in Fig. 3). Planar(A) and SPECT with sagittal (B), axial (C), and MIP (D) images are shown.

Fig. 5. Sagittal CT (A), PET (B), and fused (C) images from NaF PET/CT showing increased uptake in multipleosteophytes in the lower thoracic and upper lumbar spine and in L5/S1 degenerative disc disease (arrow).

Bone Scanning in Routine Clinical Practice 323

Fig. 6. Ability of NaF PET/CT to detect tiny lytic metastases. Focal uptake (dotted arrow) is seen in a 3-mm lyticlesion (solid arrow) in the left tibia in a patient with extensive metastatic breast cancer.

Abikhzer et al324

and oranges with oranges, not apples with oranges.Until such studies are performed, it can be arguedthat the advantages of NaF PET/CT, or at leastpart of them, are simply because they have beencompared with planar BS with or without limited-FOV SPECT. It remains to be determined whether,and to what extent, the reported advantages ofNaF PET are related to the tomographic acquisition,the higher spatial resolution and sensitivity of the

scanners, or whether there are biologic advantagesof the NaF tracer (Boxes 2 and 3).Review of published studies comparing BS and

NaF PET demonstrate the limitations when onlyplanar BS is performed and the added value ofeven single-FOV SPECT. One early studydemonstrated the superiority ofNaF-PET over planarTc-MDP BS in patients with a wide variety of malig-nancies. NaF PET detected twofold more lesions

Box 1Clinical indications for BS and/or NaF PET/CT

Benign bone disease

� Metabolic bone disease

� Osteomyelitis

� Sacroiliitis and ankylosing spondylitis17

� Arthritis

� Avascular necrosis

� Osteonecrosis36

� Paget disease5

Orthopedic applications

� Heterotopic ossification

� Painful prosthetic joints37

� Trauma and overuse injuries

� Bone grafts38

Malignant bone disease

� Primary bone malignancies39

� Metastatic bone disease23,31,40–42

Pediatric bone diseases

� Child abuse43

� Back pain17,44

Box 2Advantages and limitations of BS

Advantages

� Tc-MDP

� Wide availability

� Generator produced

� Longer physical half-life

� Lower radiation dosimetry

� Flow and blood pool studies (which oftenprovide important clinical information)

� Favorable current cost-effectiveness ratios

Limitations

� Inferior spatial resolution and sensitivity ofgamma cameras

� Longer uptake time

� SPECT, and mainly WB SPECT is not routinelyused

� CT to improve specificity is not routinely used

Box 3Advantages and limitations of NaF PET/CT

Advantages

� Sensitive for lytic lesions

� Better resolution and sensitivity of imagingdevice

� Routine tomographic acquisition

� Routine use of CT improves specificity

� Shorter total examination time

� Accurate quantitation

� Superior pharmacokinetics

Limitations

� Cost

� Cyclotron produced

� Short tracer half-life for study logistics

� Higher radiation exposure

� Lack of flow and blood pool studies

Bone Scanning in Routine Clinical Practice 325

than Tc-MDP,most notably in the spine and pelvis.40

In a prospective study by the same group in patientswith lung cancer, NaF PET detected 12 patients withmetastatic bone disease, whereas planar Tc-MDPBSdetected only 6 patients. Patients also underwentcervico-thoracic and thoraco-lumbar SPECT, whichdetected an additional 5 patients with metastaticbone disease, resulting therefore in only 1 false-negative BS study. Although the extent ofmetastatic disease was better assessed with NaFPET, this did not lead to any changes in patientmanagement. Theareasunder the receiver operatingcharacteristic (ROC) curves of NaF PET and Tc-MDPBS with SPECT were not statistically significantlydifferent, leading the investigators to conclude thatBS with limited-FOV SPECT is a practical and cost-effective examination.41 In a larger study from thesame group, including 103 patients with lung cancer,33 patients had bone metastases. There were 13false-negative planar Tc-MDP BS, 4 with Tc-MDPplanar and SPECT of the entire vertebral column,and 2 with NaF PET. Clinical management changedin 7 patients following the additional acquisition ofSPECT and in 8 with NaF PET.42

Even-Sapir and colleagues23 performeda prospective study in 44 patients with prostatecancer that aimed to determine the optimalmodality for diagnosis of bone metastases thatincluded a subanalysis of a group of patientswho underwent multi-FOV SPECT. Tc-MDP planarBS and NaF PET/CT were performed in allpatients. Twenty of the 44 patients undergoingBS also had single-FOV SPECT and 24 had

Abikhzer et al326

multi-FOV SPECT of the entire axial skeleton. Theaddition of multi-FOV SPECT improved the sensi-tivity of planar BS from 69% to 92% in the patient-based analysis, detecting 4 additional patientswith bone metastases. On the lesion-based anal-ysis, sensitivity also increased from 39% to 71%when multi-FOV SPECT was added to planar BS.Specificity in the patient-based analysis increasedfrom 64% to 82% using planar and multi-FOVSPECT, respectively, and in the lesion-based anal-ysis from 79% to 85%. NaF PET/CT sensitivity andspecificity was 100%, using both patient-basedand lesion-based analysis. Therefore, in this study,the sensitivity of 100% of NaF PET/CT in patientswith prostate cancer was 8% higher than multi-FOV SPECT. Only 1 of 13 patients with bonemetastases was not detected by multi-FOVSPECT, whereas the NaF study was positive.

Cost Effectiveness

Few cost-effectiveness analyses have been per-formed to date, all for NaF PET or PET/CT in meta-static disease. Hetzel and colleagues42 comparedNaF PET and BS (with and without SPECT) inpatients with lung cancer. Cost analysis was basedon direct costs (in Euros, V) of the procedures inGerman hospitals in 2002. At that time, the costwas 193.30 V for planar BS, 103.10 V for SPECTand 515.40 V for NaF PET. Three different cost-effectiveness strategies were analyzed. Strategy 1consistedofBSwithSPECTperformedonly incaseswith inconclusive lesions in the vertebral column onplanar scintigraphy. Strategy 2 consisted of BSand SPECT in all patients. Finally, strategy 3 con-sisted of NaF PET in all patients. A total of 87.4%,96.1%, and 98.1% of patients were correctly diag-nosed with bone metastases, for each strategyrespectively. The average cost-effectiveness ratioin the first strategy was 252 V. The incrementalcost-effectiveness ratio for strategy 2 was 1272 Vowing to additional costs of 111 V per patient. Theaverage cost per patient in strategy 3was 526V, re-sulting in an incremental cost-effectiveness ratio of2861 V compared with strategy 1 and 10,016 Vcompared with strategy 2. Considering that therewere 13 false-negative BS, 4 with BS and SPECTand 2 with NaF PET, this resulted in a change inmanagement of 9 patients with SPECT and 11 withNaF PET when compared with planar BS alone.The investigators concluded that the routine perfor-mance of SPECT was the most cost-effectivestrategy in detecting bone metastases from lungcancer.Detectionofbonemetastases in 2additionalpatients onlywithNaFPETcannot justify the twofoldincrease in costs using this diagnostic examination.The investigators furthercalculated that if thepriceof

NaF PETwas to fall to 345V or less, the incrementalcost-effectiveness ratio of NaF PET would be lessthan 1272 V and would thus be preferable to theproposedstrategyof routinely usingBSwithSPECT.In a meta-analysis on the use of NaF PET and

PET/CT for the assessment of metastatic bonedisease, a cost-effectiveness analysis based onthe overall accuracy of the 11 studies was per-formed. Using Current Procedural Terminology(CPT) rates for NaF PET or PET/CT, the averagecost per study was $1000 to $1500 (US$) witha cost-effectiveness ratio of $1038 to $1558. ForBS, including planar and SPECT, the average costper study based on Centers for Medicaid andMedicaid Services rates for 2010 was $297. Theaverage cost-effectiveness ratio was calculated at$404.16 With shortage in supply of Tc-99m associ-ated with rising costs, cost-effectiveness studieswill need to be reconsidered and updated in thefuture. It should be mentioned, however, that theadditional costs of subsequent treatments thatmay be altered or avoided owing to the superiorityof NaF PET/CT were not analyzed in any study.

SUMMARY

Two of the most common agents used for boneimaging in nuclear medicine have been reviewedand compared. Tc-MDP with gamma camerashas long been used, and physician familiaritywith this tracer is excellent. Cost-effectivenessstudies and practicality of use are currently in favorof Tc-MDP. On the other hand, NaF is a veryattractive tracer with many advantages and withthe potential to become the gold standard in func-tional bone imaging. Certain limitations remain,with routine flow and blood pool imaging not yetfeasible. There are only a few studies that have as-sessed the use of NaF in benign applications andmore are needed, also comparing Tc-MDP BSand NaF PET/CT. Technological advantages withSPECT/CT and WB SPECT discussed previouslymay potentially allow Tc-MDP to compete withNaF PET/CT in the next decade. Ultimately,regardless of which tracer will be adopted forroutine clinical use, we can rest assured that BShas a bright future.

REFERENCES

1. Blau M, Nagler W, Bender MA. Fluorine-18: a new

isotope for bone scanning. J Nucl Med 1962;3:

332–4.

2. Collier D, Fogelman I, Rosenthall L, editors. Skeletal

nuclear medicine. St Louis (MO): Mosby-Year Book;

1996. p. 4–5, 25.

Bone Scanning in Routine Clinical Practice 327

3. Ben-Haim S, Israel O. Breast cancer: role of SPECT

and PET in imaging bone metastases. Semin Nucl

Med 2009;39(6):408–15.

4. Fogelman I. Skeletal uptake of disphosphonate:

a review. Eur J Nucl Med Mol Imaging 1980;5:473–6.

5. Blake G, Park-Holohan S, Cook G, et al. Quantitative

studies of bone with the use of 18F-fluoride and

99mTc-methylene disphosphonate. Semin Nucl

Med 2001;31:28–49.

6. Wootton R, Dore C. The single-passage extraction of

18F in rabbit bone. Clin Phys Physiol Meas 1966;7:

333–43.

7. Czernin J, Satyamurthy N, Schiepers C. Molecular

mechanisms of bone 18F-NaF deposition. J Nucl

Med 2010;51:1826–9.

8. Grant FD, Fahey FH, Packard AB, et al. Skeletal PET

with 18F-fluoride: applying new technology to an old

tracer. J Nucl Med 2008;49:68–78.

9. Dworkin H, Moon N, Lessard R, et al. A study of the

metabolism of fluorine-18 in dogs and its suitability

for bone scanning. J Nucl Med 1966;7:510–20.

10. Donohoe K, Brown M, Collier B, et al. SNM procedure

guideline forbonescintigraphy:Version3.0,approved

6/20/2003. Available at: http://interactive.snm.org/

docs/pg_ch34_0403.pdf. Accessed April 19, 2012.

11. Segall G, Delbeke D, Stabin M, et al. SNM practice

guideline for sodium 18F-fluoride PET/CT bone

scans 1.0. J Nucl Med 2010;51(11):1813–20.

12. Blau M, Ganatra R, Bender M. 18F-fluoride for bone

imaging. Semin Nucl Med 1972;2:31–7.

13. Madsen M. Recent advances in SPECT imaging.

J Nucl Med 2007;48:661–73.

14. Brenner W, Vernon C, Muzi M, et al. Comparison of

different quantitative approaches to 18F-Fluoride

PET scans. J Nucl Med 2004;45:1493–500.

15. Doot R, Muzi M, Peterson L, et al. Kinetic analysis of

18F-Fluoride PET images of breast cancer bone

metastases. J Nucl Med 2010;51:521–7.

16. Tateishi U, Morita S, Taguri M, et al. A meta-analysis

of 18F-Fluoride positron emission tomography for

assessment of metastatic bone tumor. Ann Nucl

Med 2010;24:523–31.

17. Lim R, Fahey F, Drubach L, et al. Early experience

with fluorine-18 sodium fluoride bone PET in young

patients with back pain. J Pediatr Orthop 2007;27:

277–82.

18. Xia T, Alessio A, De Man B, et al. Ultra-low dose CT

attenuation correction for PET/CT. Phys Med Biol

2012;57:309–28.

19. Huang B, Wai-Ming Law M, Khong PL. Whole-body

PET/CT scanning: estimation of radiation dose and

cancer risk. Radiology 2009;251:166–74.

20. Douglas R, Cherry S. Small animal pre-clinical

nuclear medicine instrumentation and methodology.

Semin Nucl Med 2008;38:209–22.

21. Soret M, Bacharach SL, Buvat I. Partial-volume effect

in PET tumor imaging. J Nucl Med 2007;48:932–45.

22. Eriksson L, Townsend D, Conti M, et al. Potentials for

large axial field of view positron camera systems.

IEEE Nuclear Science Symposium Conference

Record. October, 2008. p. 1632–6.

23. Even-Sapir E, Metser U, Mishani E, et al. The detec-

tion of bone metastases in patients with high-risk

prostate cancer: TC99m-MDP planar bone scintig-

raphy, single- and multi-field-of-view SPECT, F18-

fluoride PET, and F18-fluoride PET/CT. J Nucl Med

2006;47:287–97.

24. Kwee T, Takahara T, Katahira K, et al. Whole-body

MRI for detecting bone marrow metastases. PET

Clin 2010;5:297–309.

25. Iagaru A, Mittra E, Yaghoubi S, et al. Novel strategy

for a cocktail 18F-Fluoride and 18F-FDG PET/CT

scan for evaluation of malignancy: results of the

pilot-phase study. J Nucl Med 2009;50:501–5.

26. Lin F, Rao J, Mittra E, et al. Prospective comparison

of combined 18F-FDG and 18F-NaF PET/CT vs. 18F-

FDG PET/CT imaging for detection of malignancy.

Eur J Nucl Med Mol Imaging 2012;39:262–70.

27. Niederkohr R. Technical feasibility vs. clinical utility:

a question of “can we?” vs. “should we?” Eur J

Nucl Med Mol Imaging 2012;39:260–1.

28. Matsunari I, Kanayama S, Yoneyama T, et al.

Myocardial distribution of 18F-FDG and 99mTc-ses-

tamibi on dual-isotope simultaneous acquisition

SPET compared with PET. Eur J Nucl Med Mol

Imaging 2002;29:1357–64.

29. Slart R, Bax J, de Boer J, et al. Comparison of

99mTc-sestamibi/18FDG DISA SPECT with PET for

the detection of viability in patients with coronary

artery disease and left ventricular dysfunction. Eur

J Nucl Med Mol Imaging 2005;32:972–9.

30. Gnanasegaran G, Cook G, Adamson K, et al.

Patterns, variants, artifacts, and pitfalls in conven-

tional radionuclide bone imaging and SPECT/CT.

Semin Nucl Med 2009;39:380–95.

31. Even-Sapir E, Metser U, Flusser G, et al. Assess-

ment of malignant skeletal disease: initial experi-

ence with 18F-fluoride PET/CT and comparison

between 18F-fluoride PET and 18F-fluroide PET/CT.

J Nucl Med 2004;45:272–8.

32. Kawaguchi M, Tateishi U, Shizukuishi K, et al. 18F-

fluoride uptake in bone metastasis: morphologic

and metabolic analysis on integrated PET/CT. Ann

Nucl Med 2010;24:241–7.

33. Beheshti M, Vali R, Waldenberger P, et al. Detection

of bone metastases in patients with prostate cancer

by 18F fluorocholine and 18F fluoride PET–CT:

a comparative study. Eur J Nucl Med Mol Imaging

2008;35:1766–74.

34. Cook G, Parker C, Chua S, et al. 18F-fluoride PET:

changes in uptake as a method to assess response

in bone metastases from castrate-resistant prostate

cancer patients treated with 223Ra-chloride (Alphar-

adin). EJNMMI Res 2011;1:4–9.

Abikhzer et al328

35. Ritt P, Vija H, Hornegger J, et al. Absolute quantifica-

tion in SPECT. Eur J Nucl Med Mol Imaging 2011;38:

S69–77.

36. Dasa V, Adbel-Nabi H, Anders M, et al. F-18 fluoride

positron emission tomography of the hip for osteo-

necrosis. Clin Orthop Relat Res 2008;466:1081–6.

37. Sterner T, Pink R, Freudenberg L, et al. The role of

[18F]fluoride positron emission tomography in the

early detection of aseptic loosening of total knee ar-

throplasty. Int J Surg 2007;5:99–104.

38. Berding G, Burchert W, van den Hoff W, et al. Eval-

uation of the incorporation of bone grafts used in

maxillofacial surgery with [18F]fluoride ion and

dynamic positron emission tomography. Eur J Nucl

Med Mol Imaging 1995;22:1133–40.

39. Brenner W, Bohuslavizki K, Eary J. PET imaging of

osteosarcoma. J Nucl Med 2003;44:930–42.

40. Schirrmeister H, Guhlmann A, Elsner K, et al. Sensi-

tivity in detecting osseous lesions depends on

anatomic localization: planar bone scintigraphy

versus 18F PET. J Nucl Med 1999;40:1623–9.

41. Schirrmeister H, Glatting G, Hetzel J, et al. Prospec-

tive 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:

1800–4.

42. Hetzel M, Arslandemir C, Konig H, 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;

18:2206–14.

43. Drubach L, Johnston P, Newton A, et al. Skeletal

trauma in child abuse: Detection with 18F-NaF PET.

Radiology 2010;255:173–81.

44. Ovadia D, Metser U, Lievshitz G, et al. Back pain

in adolescents: assessment with integrated 18F-

fluoride positron emission tomography-computed

tomography. J Pediatr Orthop 2007;27:277–82.