Nucl Med Mol Imaging Research Articles

EANM-ESR white paper on multimodality imaging

Dear readers: In the August 2007 issue of our journal, an article entitled “White paper of the European Association of Nuclear Medicine (EANM) and the European Society of Radiology (ESR) on multimodality imaging” was published. The paper was a result of working party negotiations undertaken by EANM and ESR delegations on the present and future of multimodality imaging. The issue has generated debate within the respective organizations with regard to the identity and boundaries of the specialties, professional practice, and training requirements. In this era of rapid technological advances in medical imaging that are contributing to the transformation of medicine, this debate is necessary to define the future of our specialty and its role in improved healthcare delivery. In this issue of the journal, two articles reflecting different opinions are presented. As Editor, I hope the reading of these controversy papers will help us to reflect on this important topic and to better define our personal positions and the society policies as well. Eur J Nucl Med Mol Imaging (2008) 35:673 DOI 10.1007/s00259-008-0725-1

Bone scintigraphy in electrocution wounds and comparison with FDG-PET images

One can follow the order of events that happened to this unfortunate man: The cable hits the shoulder and slides onto his back and left hip, while his right hip makes contact with a metal bar, allowing the current to leave his body (arrows). Both hands bear the signs of his attempts to throw the cable away. Positron emission tomography/computerized tomography (PET–CT) was performed to evaluate bone involvement and to provide better anatomic repair. Thus, we could exclude osteomyelitis and confirm the diagnosis of myositis ossificans caused by electrical burns (d and e). Our patient survived and could leave the hospital after treatment consisting of several autograft operations. Eur J Nucl Med Mol Imaging (2008) 35:448 DOI 10.1007/s00259-007-0629-5

Diffuse venous malformation within the left side of the body visualized by Tc-99m-labeled RBC scan

We present an unusual case of diffuse venous malformation involving the left side of the body in a 15-year-old male patient who was referred to us for evaluation of a vascular cutaneous malformation. A whole-body, Tc-99m-labeled, RBC scan showed prominent uptake within the left side of the body, involving the left upper and lower limbs as well as the anterior and posterior chest fields, with a sharp delineation in the midline of the trunk (Figure). Eur J Nucl Med Mol Imaging (2007) 34:1888 DOI 10.1007/s00259-007-0546-7

The new EANM paediatric dosage card — does it conform to ALARA for PET/CT?

Dear Sir, We read with interest the recent article by M. Lassmann and colleagues [1] on behalf of the EANM Dosimetry and Paediatrics Committees regarding the introduction of a condensed and revised version of the Jacobs et al. [2] dosage card for the major nuclear medicine diagnostic procedures. Whilst we appreciate the authors' statement that the dosage card ‘summarizes the views’ of the Committee and ‘should be taken in the context of “good practice” of nuclear medicine and local regulations’, we would like to comment on the recommended injected activities in the context of Paediatric F FDG PET. The new EANM paediatric dosage card recommends a minimum activity of 70 MBq in F FDG (2D) and F FDG (3D) PET, the latter recommended in children. At our institution we currently scale paediatric doses for F FDG PET according to weight, calculated from a baseline activity of 250 MBq for brain imaging and 350 MBq for body imaging, with a minimum administered activity of 35 MBq. Firstly, the new recommendations make no distinction between brain and half/whole body imaging either in children or adults. Secondly, on the basis of our current administered activities in the imaging of a paediatric brain, a child would have to weigh 20 kg before reaching the EANM minimum recommended activity of 70 MBq. Similarly, in body imaging a child would be required to weigh 14 kg before receiving the minimum recommended activity. These results would appear to be contrary to the ALARA principle. For example, a 10 kg child at our institution would receive 36 MBq for imaging of the brain and 50 MBq for imaging of the body. In both cases, the EANM recommended activity would be the minimum 70 MBq, (because the baseline activity x multiple=2.71× 14.0=37.94 MBq). This results in a 49% dose reduction in the brain protocol and 29% in the body protocol. These differences become greater as the child becomes smaller. We have not experienced any clinical problems using our current injected activity protocol. A recent audit of imaging performed at our institution in children aged 16 or younger over the last 2 years revealed that 28 of 150 children weighed less than 14 kg (for body imaging) or less than 20 kg (for brain imaging). All scans were of diagnostic quality. These changes are also in the context of brain imaging in 3D and body imaging in 2D with the body images acquired at 90 minutes post injection. There could be further reductions of injected activity if body imaging were performed in 3D resulting in further dose reduction. Perhaps as important for the child’s radiation dose is the dose from the CT attenuation correction methodology. Over the previous 2 years on the GE Discovery ST we used 0.8 sec rotation with 10 mA and 140 kVp for brain imaging, and 0.5 sec rotation 50 mA and 140 kVp for body imaging. This results in an approximate dose of 0.12 mSv (1 yo) to 0.06 mSv (15 yo) and 7 mSv (1 yo) to 4 mSv (15 yo) for brain and body imaging respectively. Currently we are using 120 kVp and are assessing whether further reduction is possible. The recent article by Fahey et al. suggests that these parameters could be lowered further to 80 kVp and 5 mA [3] if the CT is required for attenuation correction only. Although we realise that the aim of the paediatric dosage card was to give guidance relating to the radiopharmaceutical aspects of paediatric imaging, the dose from CT can be just as important a contribution to the total dose received by a child undergoing a PET-CT examination. Consensus on optimal CT imaging parameters is also warranted. The significant contribution of CT to the PET-CT examination in children makes it even more important that guidance takes into account our principles of abiding by ALARA. Eur J Nucl Med Mol Imaging (2007) 34:1881–1882 DOI 10.1007/s00259-007-0565-4

Is the clinical staging system a good choice in the staging of vulvar malignancies?

Dear Sir, We have read the article by Sergi Vidal-Sicart and coworkers [1] entitled “Validation and application of the sentinel lymph node concept in malignant vulvar tumours”, published in the March 2007 issue of the European Journal of Nuclear Medicine and Molecular Imaging. We agree with the authors that sentinel lymph node (SLN) identification permits the accurate pathological examination of regional nodes and could reduce the high morbidity of current surgical treatment in vulvar tumour patients. The results are impressive; however, there are some controversies in the explanation for the tumour stage in SLN-positive cases. The clinical stage of vulvar cancers is at least III when lymph node positivity is found. In the publication of Sergi Vidal-Sicart, however, there are ten patients listed with positive SLN, classified as stage Ib or II. During the past 5 years our team has been involved in the development and application of the method of SLN sampling in vulvar neoplasms to determine the level of progression of the disease. Our preliminary results suggest that sentinel node scintigraphy is a reliable method in early vulvar carcinoma and melanoma [2]. An occasional survey was written by us on the comparison of the TNM and clinical (FIGO) staging systems in different gynaecological malignancies. In 1969 FIGO introduced a staging system for vulvar cancer based on the TNM classification, utilising the clinical assessment of tumour size, tumour site, status of the regional lymph nodes and a limited search for distant metastases. The most serious drawback of this staging system was the inaccuracy of clinical assessment in determining groin node status, especially given that microscopic metastases may be present in clinically normal nodes and cancer-free nodes may be enlarged by inflammation [3]. The TNM system is a dual system distinguishing between a clinical (pretreatment) classification (TNM or cTNM) and a pathological (post-surgical histopathological) classification (pTNM). Thus, in TNM, each patient is staged clinically as well as pathologically. We should realise that not only surgical and pathological, but also clinical staging is important. We should continue to rely upon clinical staging when planning treatment and making prognostic assessments [4]. In the case of vulvar carcinoma, the TNM staging system is obviously superior to clinical (FIGO) staging. Of all the gynaecological malignancies, vulvar carcinoma is the one for which the SLN concept is most relevant. We suggest that in publications dealing with SLN investigations, TNM staging should be preferred over clinical staging. Eur J Nucl Med Mol Imaging (2007) 34:1878–1879 DOI 10.1007/s00259-007-0511-5

The “White paper of the European Association of Nuclear Medicine (EANM) and the European Society of Radiology (ESR) on multimodality imaging”: a message from the EANM and ESR Presidents

While nuclear medicine is growing, new areas emerge where relations with other medical specialties are necessary to deliver appropriate patient care. Cardiovascular nuclear medicine in the recent past, and now multimodality imaging provide clear examples. Fused and hybrid imaging with SPECT/CT, PET/CT, and other modalities represent an area of very rapid growth with important professional implications. In 2005, the executive committee of the European Association of Nuclear Medicine (EANM) started formal contacts with the leadership of the European Society of Radiology (ESR) to discuss areas of common interest, to define forms of potential cooperation, and to advance multimodality imaging in medicine. In particular, both associations recognised the importance of coordinating working practices for such imaging systems and that undertaking the nuclear medicine and radiology components of imaging with hybrid systems requires different skills. To properly address this important issue, a strategy committee was called to give advice to the executive committee on how to develop the relations with the ESR. At the same time, the EANM executive committee gathered advice from national delegates and experts. Following the first contacts between the leadership of both societies at the European Congress of Radiology (ECR) 2005 in Vienna, an EANM/ESR joint ad hoc committee was formed, comprising the president, the president elect and the president of the UEMS section of each society. This committee had a series of meetings to discuss training, accreditation and professional issues, and to plan educational activities. In the meantime, joint sessions during the EANM and ECR congresses have been initiated to highlight the clinical value of SPECT/CT and PET/CT in comparison with other existing imaging modalities. In this issue of the European Journal of Nuclear Medicine and Molecular Imaging, a document entitled “White paper of the European Association of Nuclear Medicine (EANM) and the European Society of Radiology (ESR) on multimodality imaging” is published [1]. This paper represents one of the results of the work that has been done by the joint ad hoc committee during the last two years and sets out the positions and aspirations of the EANM and the ESR to working together on an equal and constructive basis for the future benefit of both specialties. It was approved by the executive committee of the EANM, the executive council of the ESR, the general assembly of the European Association of Radiology and the executive committees of the UEMS sections of nuclear medicine and radiology. The document was presented at the EANM strategy committee meeting held in London in February 2007. It was also presented at both the EANMadvisory council meeting and the EANMextraordinary delegates’ meeting that took place in Vienna in March 2007. While it is important to underline that the white paper is not a formal guideline to be followed by all national societies, it is hoped that it will stimulate discussion in all Eur J Nucl Med Mol Imaging (2007) 34:1145–1146 DOI 10.1007/s00259-007-0495-1

Intense [18F]FDG tongue uptake in a case of acipimox-related angio-oedema during FDG-PET myocardial viability study

Tongue FDG uptake can occasionally be seen during FDGPET imaging, especially in patients talking after injection [1]. Orofacial angio-oedema can be triggered by various drugs [2]. Acipimox, a nicotinic acid analogue, used in FDG-PET myocardial viability studies can rarely cause angio-oedema [3]. A kidney mass suspicious for malignancy was found on MRI scan in a 74-year-old man with haematuria and coexistent advanced coronary artery disease. Myocardial perfusion assessment for pre-operative cardiac risk stratification further necessitated an FDG-PET viability study, which, because of the patient’s history, was combined with simultaneous whole-body scan. Acipimox was orally given before the administration of 460 MBq [F]FDG [3]. One hour after acipimox intake, the patient reported difficulty in swallowing. After scan completion, neck, tongue, tonsils and soft palate oedema was evident (a). Elevated total IgE (311 IU/ml; normal<100 IU/ml) was compatible with IgE-mediated angio-oedema. Whole-body scan was negative for malignancy. However, intense uptake (SUV maximum 8, average 6.2) in the tongue and mild activity along the pharynx and upper oesophagus were best shown in sagittal sections (b). Activity was located only at the posterior, swallowing-related and not at the anterior, freely moving part of the tongue. The area of activity was lacking the normal upward curved shape of the tongue, probably because of the swollen soft palate (c, arrow). In the presented case, intense FDG uptake of the tongue was shown in association with acipimox-related angiooedema. Increased muscular metabolism due to repeated swallowing efforts, triggered by the swollen soft palate, is proposed as the underlying mechanism. Eur J Nucl Med Mol Imaging (2007) 34:1320–1321 DOI 10.1007/s00259-007-0468-4

PET probes and oncological surgery: a productive new marriage for nuclear medicine?

Surgical excision is still the best option in attempting to cure many solid cancers. Even patients with tumours that have already metastasised to regional lymph nodes or to distant sites are potential candidates for surgery [1]. With this in mind, accurate staging of the disease and detection of all cancer sites are of the utmost importance in order to allow selection of those patients who may really benefit from complete surgical resection. Several imaging modalities are currently used to preoperatively assess cancer patients. Among them, over the past decade functional imaging with positron emission tomography (PET) has emerged from being primarily a research tool to become a valuable non-invasive technique which is now playing an increasingly important role in the diagnosis and staging of malignant disease, treatment monitoring and therapy planning. In particular, PET with the labelled glucose analogue F-fluorodeoxyglucose (FDG) has become an established modality for cancer imaging. However, this technique is not perfect: small lesions and those adjacent to sites of high physiological FDG uptake can be missed; moreover, false-positive results due to benign diseases are possible [2]. The introduction into clinical practice of hybrid PET/computed tomography (CT) scanners has solved some of these problems: fused PET/CT imaging allows accurate anatomical localisation of functional abnormalities and reduces the incidence of falsepositive findings by facilitating the differentiation of normal physiological uptake from pathological uptake. Many recently published studies have demonstrated that, in several clinical applications, PET/CT imaging consistently outperforms PET and CT imaging carried out separately and analysed together [3]. Nevertheless, the suboptimal spatial resolution of PET still limits the accuracy of PET/CT in characterising lesions ≤5 mm. It is intuitively expected that the availability of a PETsensitive probe able to intra-operatively identify all FDGavid tumours may improve preoperative PET staging because the detection system of such a device can be placed close to the radiation source [1]. Intra-operative probes have become important devices in nuclear medicine, especially the gamma probes which detect gamma photons from single-photon emitters like Tc and In [4, 5]. In fact, in addition to the inspection and palpation traditionally used by the surgeon, intra-operative detection of radiopharmaceuticals by hand-held gamma probes is increasingly been performed. The main clinical applications of radioguided surgery are sentinel lymph node mapping for several kinds of tumour and detection of parathyroid adenoma [6]. The use of an intra-operative probe during surgical oncology, after the administration to the patient of a radiopharmaceutical that accumulates in the tumour, can help the surgeon in finding the neoplastic lesions. Moreover, by means of the probe, the surgeon can also confirm that the excised parts really do contain cancer tissue, and that the tumour has been completely removed. In this application, the positron emitter FDG can be detected by two methods: detection of the beta particles or detection of the annihilation gamma photons. In this issue of the European Journal of Nuclear Medicine and Molecular Imaging, Piert and colleagues [7] report the results of a beta and gamma probe system for the intra-operative localisation of tumour sites in 17 patients Eur J Nucl Med Mol Imaging (2007) 34:1530–1533 DOI 10.1007/s00259-007-0482-6

PET/CT demonstrates increased myocardial FDG uptake following irradiation therapy

Late myocardial damage induced by radiotherapy has become an important issue in radio-oncology since several studies have demonstrated increased incidence of cardiovascular disease following radiotherapy [1–3]. A 52-year-old man underwent radiochemotherapy and resection of oesophageal cancer in 2002. In June 2006, [F]fluorodeoxyglucose positron emission tomography (FDG-PET)was performed for staging of newly detected oral cancer. While FDG-PET was unsuspicious in the head and neck area, increased FDG uptake in the mediastinal myocardium with unusual sharp borders excluding the apex was observed. As shown in the figure, superimposition of radiation dose distribution to the PET/CT revealed that increased FDG uptake was restricted to parts of the myocardium that were irradiated to more than 35 Gy. The patient had no clinical symptoms of angina pectoris or heart failure and had a normal ECG. Perfusion scintigraphy revealed homogeneous myocardial uptake without perfusion defects and regular global and regional myocardial contraction. The regions of high FDG uptake corresponded to the irradiated field but not to any coronary artery vascular territories. It is unlikely that these findings were due to coronary artery stenosis. Increased myocardial FDG uptake might be explained by microvascular damage due to irradiation and impairment of fatty acid metabolism which was counterbalanced by increased glucose metabolism. Alternatively, radiation may have induced microvascular damage of the endothelial cells of capillaries within the myocardium, causing capillary swelling and progressive Eur J Nucl Med Mol Imaging (2007) 34:1322–1323 DOI 10.1007/s00259-007-0469-3

FDOPA PET has clinical utility in brain tumour imaging: a proposal for a revision of the recent EANM guidelines

Dear Sir, We read with great interest the “EANM procedure guidelines for brain tumour imaging using labelled amino acid analogues” that were published in the November issue of the Eur J Nucl Med Mol Imaging [1]. The publication of an EANM guideline is a professional event of major importance, and the present most valuable work is based on an extensive survey of the literature. However, we would suggest that 6-fluoro-(F)-L-3,4dihydroxyphenylalanine (FDOPA) needs to be included in the limited guideline list of radiopharmaceuticals. FDOPA is clearly a “labelled amino acid analogue” and is currently registered in France for brain tumour imaging, while the three other radiopharmaceuticals that are listed in the guideline [iodomethyltyrosine-(I) or IMT, methionine(C) or MET or fluoroethylthyrosine-(F) or FET] have no marketing authorisation. We would like to briefly summarise the published data that support the use of FDOPA PET in brain tumours and thus favour a limited revision of the guidelines. The first description of imaging of a brain tumour with FDOPA PET was a case report by Heiss in 1996 [2]. A 57year-old woman suffering from a minor movement disorder of the left arm and hand was referred for FDOPA PET. The FDOPA PET study not only proved asymmetrically reduced dopamine uptake in the putamen, but also revealed pathologically increased FDOPA accumulation in the right frontal lobe. Further PET examinations demonstrated increased MET uptake and low glucose metabolism in this right frontal region. Histology confirmed the diagnosis of a grade 2 oligo-astrocytoma. In 2003, Becherer et al. [3] reported on the performance of FDOPA PET in this setting, in comparison with MET, in series of 20 patients with known supratentorial brain lesions. The diagnoses were 18 primary brain tumours, one metastasis and one non-neoplastic cerebral lesion. MET and FDOPA images matched in all patients and showed all lesions as hot spots with higher uptake than in the contralateral brain. Standardised uptake value ratios for the tumour to the contralateral side (mean±SD) were similar for MET and FDOPA: 2.05±0.91 vs 2.04±0.53. The benign lesion, which biopsy revealed to be focal demyelination, was false positive, showing increased uptake of both MET and FDOPA. Recently, Chen et al. [4] compared the diagnostic performance of FDOPA PET and FDG PET in 30 patients with brain tumours: seven at diagnosis and 23 in whom recurrence was suspected. Visual analysis, considering as malignant any focus with an intensity greater than that of background, revealed a greater sensitivity for FDOPA (22/23=96% vs.14/23=61% for FDG), with an identical specificity: 3/7=43%. In fact, this low specificity was due to three cases with post-irradiation necrotic areas that faintly took up FDOPA to a much lesser extent than viable tumour tissue but that nevertheless had to be considered false positive results at visual analysis. To solve this problem, the authors proposed use of the tumour/striatum activity ratio, with a cut-off value of 0.75. The validity of this proposal was tested in another series of 51 patients including 47 with Eur J Nucl Med Mol Imaging (2007) 34:1131–1132 DOI 10.1007/s00259-007-0400-y

Data acquisition and analysis: the strength of methodology in nuclear medicine and molecular imaging

Dear Sir, We read with great interest the paper “Data acquisition and analysis: the strength of methodology in nuclear medicine and molecular imaging” by G. Lucignani [1]. The author nicely demonstrates the need for image quantification of positron emission tomography (PET) studies. Typically a PET acquisition consists of an emission scan and a transmission scan. Iterative reconstruction methods yield a better result compared with the filtered back projection method, especially in terms of noise. However, iterative reconstruction methods have a tendency to give rise to noise artefacts when a large number of iterations are performed. Image quality also depends greatly on how correction methods, such as attenuation and scatter correction, are applied, because these methods may introduce additional statistical noise into the reconstructed data sets. The complexity of different reconstruction methods, i.e. different correction methods, and the variation in scanner characteristics make it very difficult to compare the results between institutions. The effect of reconstruction on NEMA contrast values was demonstrated recently [2], and an interlaboratory comparison study of image quality using the NEMA NU 2-2001 procedure for assessment of image quality showed significant differences between different PET scanners [3]. The systematic difference in contrast entails two practical problems: visual image interpretation of scans may become more and more difficult and a systematic difference in measured SUVs may affect the diagnostic accuracy to be compared inter-institutionally. Using a common whole-body PET study and differing OSEM reconstruction parameters (number of iterations and subsets) of the emission data sets, we found no significant differences in the maximum SUV, but there were differences of up to 30% in the homogeneity of the liver. Of course, the SUV has an important clinical significance, but the homogeneity of a study is important for the clinician as well. Changing the reconstruction parameters of transmission data sets did not have any significant impact, because we used a segmented attenuation correction method. This has also been found to reduce the metal artefacts in PET-CT studies [4]. If the mathematical algorithm fails to perform an accurate attenuation correction, the image quality is degraded. Applying a phantom study by varying the transmission acquisition times (600, 300, 200, 100 or 20 s) and using Ge rods for attenuation correction, we observed sudden changes of up to 23% in homogeneity, but very minor differences in maximum SUV. Our preliminary findings show that in order to achieve comparable results of clinical studies from different institutions, the specification of both emission and transmission acquisition parameters has to be as detailed as possible. Eur J Nucl Med Mol Imaging (2007) 34:961–962 DOI 10.1007/s00259-007-0399-0

W. Mohnike and G. Hör: PET/CT Atlas—ein interdisziplinärer Leitfaden der onkologischen PET/CT Diagnostik (PET/CT Atlas—an interdisciplinary guide for PET/CT diagnosis in oncology)

Experts all over the world agree that PET/CT is one of the most rapidly emerging imaging modalities in oncology. Clinical acceptance has been accelerated by the fact that the combined metabolic–morphological fusion imaging offered by PET/CT scanners overcomes the lack of lesion localisation on PET alone. For many years, clinicians have been used to morphological imaging, and before nuclear medicine offered this combined imaging modality they had difficulty in accepting PET imaging alone for daily practice. This book, written in German, covers most aspects of PET/CT imaging in oncology. It is divided into 15 chapters, including an introduction to the topic, a chapter on the basics of PET and PET/CT, 11 clinical chapters, one chapter on radiotherapy planning and a concluding chapter on future aspects of molecular PET. Wolfgang Mohnike from the Diagnostic Centre of Berlin and Gustav Hor, who was Director of the Centre of Radiology at the J.W. Goethe University of Frankfurt, provide an excellent atlas on PET/ CT for nuclear medicine physicians, radiologists and also our clinical colleagues who are interested in modern imaging modalities in oncology. The introduction starts with a historical overview followed by different aspects of PETand PET/CT, including the basics of physics and biochemistry, cost effectiveness, technical developments, indications for PET and PET/CT, and the influence of PET/CT on the increase in PET investigations; it also gives an insight into the PET screening programmes in Japan and Taiwan. The second chapter, on the physical and technical basics of PET and PET/CT, is excellently written by Thomas Beyer, a well-known pioneer of PET/CT, who, together with David Townsend, developed the first prototype of this combined imaging modality. This chapter provides important information about PET radioisotopes and tracers and offers a clear and well-illustrated insight into the technical principles of PET and PET/CT. It also covers the differences between the commercially available PET/CT scanners, the advantages and disadvantages of different crystals, imaging protocols, the arguments for and against the use of intravenous and oral contrast media, pitfalls and possible artefacts, and radiation protection aspects. Chapters 3–12 describe the use of PET/CT in the most accepted indications in oncology, including (in order) lung, gastroenterological, gynaecological, urological, head and neck and thyroid cancers, malignant melanoma, lymphoma, bone cancers, paediatric cancers and carcinoma of unknown primary, and chapter 13 covers miscellaneous aspects. Each chapter starts with a clinical introduction, including epidemiology, pathogenesis, treatment options, conventional imaging and other information, although this is not consistently done for all clinical topics, and describes accepted and special indications according to the different tumour entities. This introductory part of each clinical chapter is followed by a large number of interesting cases. In all cases an anamnesis, the PET/CT report and a conclusion are provided. The quality Eur J Nucl Med Mol Imaging (2007) 34:963–964 DOI 10.1007/s00259-007-0382-9

Radioguided localisation (ROLL) of non-palpable breast lesions and simultaneous sentinel lymph node biopsy (SNOLL): the experience of the European Institute of Oncology

Dear Sir, We read with interest the recent article by J.E. Jansen and colleagues [1] regarding the influence of wire localisation for non-palpable breast lesions on visualisation of the sentinel node. We were particularly struck by their conclusion that guide wire localisation prior to the sentinel node procedure negatively influences visualisation of the sentinel node. In the reported series, a total of 136 patients underwent a wireguided tumour excision combined with a sentinel node procedure. Sixty-six patients had guide wire localisation prior to the sentinel node procedure, while 70 had sentinel node visualisation before insertion of the guide wire. The sentinel node was visualised in 41 (62%) of the patients who first underwent guide wire localisation. In the group of patients who underwent visualisation of the sentinel node before placement of the guide wire, the sentinel node was visualised in 62 (89%). This was a significant difference in visualisation (p<0.001). The authors concluded that insertion of a wire in patients with occult breast tumours somehow disturbs the subsequent sentinel node procedure. We would like to point out some considerations. First, regardless of the different timing of the sentinel node biopsy (prior to or after guide wire localisation), the sentinel lymph node identification rate in the series of Jansen et al. appears to be much lower than that reported in the current literature. In the American Society of Clinical Oncology (ASCO) guideline recommendations for sentinel lymph node biopsy in early stage breast cancer [2], 69 studies (10,454 patients) were analysed in which sentinel node biopsy was compared with completion axillary dissection. The overall sensitivity of the sentinel node biopsy ranged from 71% to 100%, but was about 95% in the larger reported studies in which the lymphoscintigraphic technique (alone or in association with the blue dye technique) was used. In the pioneering experience of the European Institute of Oncology, recently published [3], the sentinel node identification rate was 99%. Secondly, in order to explain the findings of their study, Jansen et al. affirm that the localisation of a wire somehow changes the local anatomy and physiology, resulting in a disturbance of tracer drainage. They point out that oedema caused by an inflammatory reaction as a result of the impact of insertion of the wire may offer an explanation for the changes. They also mention that wire insertion causes higher external pressure on lymphatic channels, resulting in distal obstruction of these vessels; this is cited as another possible explanation, especially when several attempts are needed to place the wire at the right location, thus increasing the local impact on the breast tissue. Unfortunately, neither a previous diagnostic biopsy (e.g. fine-needle aspiration cytology, vacuum-assisted biopsy) nor a previous excision biopsy, including tumorectomy and quadrantectomy, is considered a contraindication to sentinel node biopsy. Prior breast surgery does not modify the success rate of sentinel node identification [2], so how can a less invasive procedure such as a simple wire insertion decrease the sentinel node identification rate? Finally, we agree with Jansen et al. that radioguided occult lesion localisation (ROLL) for non-palpable breast lesions can easily substitute for the wire insertion techEur J Nucl Med Mol Imaging (2007) 34:957–958 DOI 10.1007/s00259-007-0397-2

PET/CT in head and neck cancer

Head and neck cancer is the third most frequent cancer worldwide, with 500,000 new cases every year [1]. It is a heterogeneous group of neoplasms arising in the upper aerodigestive tract with different pathophysiologies, histologies and staging paradigms [2]. PET/CT is emerging as the “method of choice” for diagnosis in head and neck cancer [3], and nuclear medicine physicians may regard this diagnostic entity as a rather simple challenge: the cancer, the large nodes and the metastases are often clearly visible on FDG PET/CT and we can just describe what we see and regard the case as closed. But we could also try to obtain a thorough knowledge of this complex and intriguing heterogeneous disease entity. We could aspire to learn more about the different types of head and neck cancer and their pathogenesis, to improve staging paradigms, to refine strategies for diagnosis and treatment, and to clarify the strengths and weaknesses of PET/CT for each type. Such an approach would facilitate better, more professional and closer collaboration with the surgeons, oncologists and radiation oncologists who are responsible for patients’ treatment. Furthermore, only through this approach, whereby we accept responsibility as clinicians ourselves, can we exploit fully the advantages of PET/CT to achieve optimal diagnosis and treatment guidance in head and neck cancer. Head and neck cancer is more common in males than in females. It is found most frequently in middle-aged and elderly patients, and smoking and alcohol consumption are involved in the pathogenesis of some of the most frequent types. Overall, head and neck cancer patients have a 5-year survival rate of only 50%, though the rate varies widely depending on the type and extent of disease, and an improved effort with regard to both diagnosis and treatment is warranted. Head and neck cancer may arise in the nose and paranasal sinuses, the lips and oral cavity, the naso-, oroand hypopharynx, and the salivary glands. Malignant melanomas and invasive skin cancers can arise in the head and neck region. Head and neck sarcomas are seen in children, most frequently as soft tissue sarcomas (rhabdomyosarcomas), and in adults as osteosarcomas. Osteosarcomas of the head and neck region are seen in a slightly older age group than the more common osteosarcomas of the extremities, and they have a lower potential for metastatic spread. In the head and neck region, rarer nerve and vessel cancers can also be found. The neck also houses the thyroid and parathyroid glands, which give rise to neoplastic diseases well known in nuclear medicine. A thorough knowledge of these cancers and their mode of local invasion and spread will lead to better and more useful PET/CT interpretation. The histological type is most often a squamous cell carcinoma owing to the site of origin, but adenocarcinomas and many other rarer histologies are found. The special non-keratinized and undifferentiated carcinomas of the nasopharynx are especially frequent in Asia, China and Japan and among the circumpolar Inuit population (owing to a combination of genetic predisposition, Epstein-Barr virus infection and environmental co-factors). Our institution (Rigshospitalet, the National University Hospital in Copenhagen) is the referral hospital for Greenland, with its Eur J Nucl Med Mol Imaging (2007) 34:1329–1333 DOI 10.1007/s00259-007-0373-x

Molecular imaging is indispensable for the development of stem cell-based myocardial regenerative therapy

Stem cells have the potential to proliferate, self-renew and differentiate into various organ-specific cell types. They can be of either human or non-human origin, and they can be classified according to (a) their developmental stage, as embryonic, fetal or adult; (b) their potential to differentiate into one or more specific types of mature cell, as totipotent, pluripotent or multipotent; and (c) the tissue of origin as haematopoietic, mesenchymal, skeletal or neural. Embryonic stem cells are totipotent and have the potential to become all specialized cell types, whereas adult stem cells are undifferentiated cells found in differentiated tissue with the potential to renew themselves and to differentiate to yield organ-specific tissues. The remarkable proliferative and differentiation capacity of stem cells has raised the expectation of an almost unlimited supply of specific cell types. It has been therefore hypothesized that stem cells could differentiate into cardiomyocytes. In their review in the Journal of Nuclear Cardiology, Chang et al. [1] from the Department of Medicine, Divisions of Cardiology and Radiology, Molecular Imaging Program at Stanford, Stanford University School of Medicine, California discuss the use of stem cell therapy and imaging in the context of cardiology and point out that stem cell therapy is emerging as a promising approach to treat heart diseases. Considerable evidence from experimental studies and initial clinical trials suggests that stem cell transplantation promotes systolic function and prevents ventricular remodelling. Thus stem cells are candidates for novel therapeutic strategies for patients with different heart diseases most commonly caused by myocardial infarction. However, the specific mechanisms by which stem cells improve heart function remain largely unknown. In addition, interpreting the long-term effects of stem cell therapy is difficult because of the limitations of conventional techniques. Recent randomized studies of cell therapy for heart disease [2–7] constitute examples of how rapidly this field is developing and they also uncover many critical issues that have yet to be addressed [8]. The fundamental concepts related to the use of stem cells in cardiovascular disease have also been summarized by Penn and Mal from the Department of Cardiovascular Medicine, The Cleveland Clinic Foundation, in their recent paper [9] and by van der Bogt et al. from the Molecular Imaging Program at Stanford, Laboratory of Cardiothoracic Transplantation, Stanford University School of Medicine, California, and the Department of Surgery, Leiden University School of Medicine, The Netherlands. Indeed, the variety of available cell types and methods for cell delivery to the myocardium raises questions about their use for therapy [10]. In his paper in Heart Failure Reviews [11], Bengel, from the Division of Nuclear Medicine, Russell H Morgan Department of Radiology and Radiological Sciences, Johns Hopkins University Medical Institutions, remarks that, despite an increasing body of experimental and human data, this innovative and promising therapeutic approach for heart failure still presents a substantial challenge for basic scientists and clinical researchers. In particular, two major concerns are the ability of undifferentiated stem cells to form teratomas and the possibility of a provoked immune reaction after transplantation of stem cells into a new host. In their paper, van der Bogt and colleagues discuss the Eur J Nucl Med Mol Imaging (2007) 34:422–425 DOI 10.1007/s00259-007-0369-6

Role of 99mTc-sestamibi SPECT in accurate selection of primary hyperparathyroid patients for minimally invasive radio-guided surgery

Dear Sir, We read with interest the recent paper by Rubello et al. entitled “Role of Tc-sestamibi SPECT in accurate selection of primary hyperparathyroid patients for minimally invasive radio-guided surgery” [1]. Minimally invasive radio-guided surgery (MIRS) is widely employed for primary hyperparathyroidism caused by a solitary parathyroid adenoma. Since an intraoperative gamma probe is used, significant sestamibi uptake by the adenoma is important [2, 3]. Obviously, as the authors underline, radiation attenuation problems may cause underestimation of the sestamibi uptake by adenomas in an anterior planar image, resulting in doubt as to the appropriateness of MIRS in some cases. In this regard the authors report an interesting finding, namely that the parathyroid to background (P/B) ratio derived from SPECT correlates well with the intrasurgery P/B ratio; this means that using SPECT it is possible to predict the intraoperative measurement obtained with the gamma probe, thus allowing better patient selection for MIRS. However, we believe that this finding is not going to have a significant impact on the selection of patients for MIRS since only patients with a high probability of having a solitary adenoma are candidates for MIRS [4]. In our experience most of these patients demonstrate a P/B ratio significantly higher than the threshold, referred to by the authors, of 1.2 in the anterior planar image and thus do not require any quantitative confirmation. In the case of deeply located adenomas, despite the relatively reduced sestamibi intensity in the anterior planar image, they will still be well recognised as parathyroid lesions and any doubt about the appropriateness for MIRS in those cases can be dispelled just by knowing that the adenoma is located deeply. In this regard we visually estimate the depth and at the same time take into account the attenuation effect on sestamibi uptake from adenomas by using oblique planar imaging in all cases, rather than the time-consuming and rather complicated SPECT technique. Oblique planar imaging is easy to perform and very valuable even when the adenoma is behind the thyroid gland [5, 6]. In the presented study the ratio derived from SPECT was useful in only 1/22 patients although 7/22 adenomas were deeply located. The quantification might possibly be useful in estimating the appropriateness of the sestamibi uptake for MIRS in a few borderline cases. However, trying to include these patients will inevitably increase the probability of multiple gland disease since it is well known that frequently only the largest gland is visible in hyperplasia, especially in the early stages, when the size of the other glands remains below the sensitivity of the techniques used [7, 8]. The result will then be that more operations will be changed to bilateral exploration assuming that rapid PTH measurement is available. In this regard, SPECT, because of its slightly better sensitivity compared with planar imaging, would be most useful not in selecting more patients but rather in excluding some on the basis of the identification of more than one parathyroid lesion in cases of multi-gland disease. Nevertheless, ultrasound can contribute this important information even better than SPECT. In conclusion, we believe that in the selection of eligible patients for MIRS, quantification is not needed and that performance of SPECT exclusively for this purpose is not justified. Eur J Nucl Med Mol Imaging (2007) 34:607–608 DOI 10.1007/s00259-006-0332-y

Is PET-CT the only option?

Accurate determination of the anatomic location of normal and pathological sites visualized on PET undoubtedly facilitates optimal interpretation of PET studies. It is now well established that image fusion with structural studies improves the diagnostic accuracy of images generated by PET [1]. PET images can be fused with CT and/or MRI either by using co-registration algorithms for PET and CT scans which have been acquired separately or by using combined PET-CT instruments. However, combined PETCT scanners have certain advantages. Sequential acquisition of PET and CT images (one immediately after the other) minimizes patient and organ motion artifacts since the patient is positioned almost identically for both scans [2]. Furthermore, the CT data can be used for attenuation correction (AC) of the PET emission scan [3, 4]. Traditionally, an external radionuclide source has been employed for this purpose, but the use of CT for AC provides several advantages: First, the transmission scanning time is much shorter for CT than for an external radionuclide source (subminute versus 20–30 min) [4]. Second, transmission scans are acquired by using external radionuclide sources with 511 keV or higher energies (Ge, Cs) which provide images with suboptimal quality for structural assessment of anatomic sites; in contrast, CT images offer high spatial resolution and low noise, facilitating structural assessment of abnormalities visualized [3, 4]. Third, by using X-rays (CT) the cost of replacing external radioactive sources is eliminated for the life of the machine [3]. In addition, combined PET-CT can help to distinguish true lesions from physiologic uptake, particularly in patients whose anatomy is distorted by postoperative or radiation therapy. Many recent studies highlight the importance of the combined units when adopted in the appropriate clinical settings [5–8]. However, PET-CT units as the “only option” may not be realistic for multiple reasons. In a major fraction of patients who undergo PET-CT scanning a diagnostic CT is performed for further assessment of the disease sites, entailing the performance of two CT scans; the radiation dose may consequently reach high levels, especially in the pediatric population. In addition, misalignment between PET and CT images due to voluntary or physiological motion of the lungs, heart, bowel, and other organs may cause artifacts in the fused image [9–12]. CT scans are acquired during breath holding while PET images are taken over 3–4 min, with averaging of data from different fields of view during this extended period. Thus, respiratory artifacts caused by different breathing patterns during the acquisitions of PET and CT data can be significant [9, 10, 13, 14]. Curvilinear cold artifacts paralleling the dome of the diaphragm at the lung/diaphragm interface have been noted on 84% of PET-CT image acquisitions but are not seen on germanium transmission source-corrected images [9]. Another technical limitation of PET-CT scans is the necessity of applying compensatory algorithms to the CT scan data to produce an equivalent degree of attenuation between CT attenuation-corrected PET/CT scans and radionuclide attenuation-corrected PET scans [15]. Therefore, precise quantitative measurement of radiolabeled Eur J Nucl Med Mol Imaging (2007) 34:819–821 DOI 10.1007/s00259-006-0340-y

FDG-PET in the clinical management of carcinoma of unknown primary with metastatic cervical lymphadenopathy: shifting gears from detecting the primary to planning therapeutic strategies

Dear Sir, With the introduction of FDG-PET in clinical oncology, one of several unresolved areas that are being investigated is detection of the primary tumor in patients with nodal metastases from an unknown site. The results, however, have indicated varying degrees of success, with the rate of detection of putative primary tumors ranging from 10% to 60%. This wide variation can be attributed to the criteria adopted to define the “unknown primary” in different clinical settings. According to the strict definition, the term “unknown primary” should be used when clinical examination, CT/MRI and panendoscopy fail to reveal a primary tumor. Unfortunately, only a few studies have adhered to such a strict definition. A careful analysis of the studies which employed these uniform criteria reveals only a modest increase in the detection rate when using FDGPET. Systematic reviews [1–4] published at various stages of evolution of PET have continued to contradict each other on this issue; the two most recent reports [1, 2] estimated the benefit to be modest at best (27% and 25% respectively) after all other modalities had failed. A head-to-head comparison of the performance of CT, PET, PET and CT side-by-side, and co-registered PET-CT documented a success rate of 25%, 25%, 29%, and 33% respectively, with no statistically significant difference among the modalities [5]. There are reports from prospective studies [6] that FDG-PET does not add significantly in finding the occult primary in a carefully worked-up CUP, especially if conventional imaging modalities are negative. In practice, it is not uncommon to find that a primary detected by FDGPET was actually a “missed primary” rather than an unknown primary in the true sense. While the results of FDG-PET in detecting an occult primary in the setting of CUP are not encouraging, an important observation which has gradually become obvious as an offshoot in several studies (including those in which FDG-PET was found to be non-contributory in detecting the primary after conventional work-up) has been the serendipitous discovery of multiple sites of disease, which has implications for the treatment strategy. By virtue of its exquisite sensitivity, FDG-PET whole-body survey can be extremely useful in guiding the therapeutic strategy by identifying or ruling out additional sites of disease in patients who present with a single metastatic site. The study by Fogarty et al. [6], which found a limited role for FDGPET in diagnosing occult primaries, concluded that the modality is of substantial benefit in the detection of unsuspected distant disease in patients with undifferentiated nodal disease and in the delineation of regional disease in patients with N2 disease. In a cohort study reported by Johansen et al. [7], whole-body F-FDG PET had treatment-related implications in 24% of patients. Finally, a systematic review [2] of 16 studies on 302 patients found that FDG-PET detected previously unrecognized metastases in 27% of patients, of which 11% were distant metastases, and resulted in changes in treatment in 24.7% of cases. This is an important consideration while planning treatment for patients with CUP with cervical node metastasis. While locoregional treatment is planned for those with one site, it is rarely considered for those with distant metastatic sites. Also, the detection of distant metastases can lead to a change in the treatment intent, from cure to palliation. Eur J Nucl Med Mol Imaging (2007) 34:427–428 DOI 10.1007/s00259-006-0313-1

Is a methodology available that accurately measures the cost of an FDG-PET study?

Health care expenses are escalating in many countries. With more restrictive reimbursement schemes based on case-mix funding systems, increasing complexity and rising costs, the health care sector faces a new challenge in becoming more cost efficient. In nuclear medicine, as in other specialties, the rapid growth and development of medical technology has contributed significantly to the steady increase in the costs of medical care. Hospital administrators and governments are being forced to take difficult decisions regarding which new developments in nuclear medicine should be supported financially, e.g. construction of a cyclotron facility, acquisition of new hybrid devices (SPECT/CT and PET/CT) or building of shielded rooms for metabolic therapies [1, 2]. Cost-effectiveness is now being used by many payers as a criterion for health care coverage, necessitating estimation of the economic effects of incorporating a technological innovation. Implicit in economic analysis is the assignment of some monetary cost to the considered intervention. Cost measurement is a complex matter, however, and there is a lack of clarity both in the terminology and in the cost concepts utilised. Cost is a term that can have several definitions in medical economics [3]. Especially in fee-for-service health systems, “costs” may refer to charges or to actual direct medical costs. A health care charge is synonymous with the administrative price, reflecting the amount requested by an institution or office practice for a given admission, procedure or office visit from third party payers or patients. Most health care organisations have a predetermined charge for each type of service, but the charge may not accurately portray the underlying true cost of providing the service because of historical and political factors, regulatory compliance, budget constraints and market forces. Allocating cost based on revenues may become arbitrary when fees are determined independently of costs [1]. Although the importance of accurate cost data in respect of an FDG-PET study has been recognised, most cost estimates have focussed on average cost per procedure extrapolated from a global cost. Because they are easy to obtain, indirect top-down methods are used to estimate the costs: (a) using hospital charges to estimate hospital costs and physician fees to estimate the cost of professional services, (b) averaging multiple hospital and physician charges, (c) adjusting charges by cost–charge ratios and (d) using third party payments as proxies for resource costs. The differences in methods used, cost components included and health care systems make comparisons unreliable [4–6]. In this era of intense cost containment there is an urgent need for more accurate data on the precise costs of a health care procedure. Unfortunately, we currently lack well-accepted metrics for the accurate determination of costs. Current hospital accounting systems do not provide the information needed to determine the cost of a medical service or to make decisions. Traditional accounting systems break down costs into direct and indirect costs. Direct costs comprise expenditure on resources, in the form of specific products (e.g. the price of radiopharmaceuticals or other consumables, the acquisition or repair cost of a PET scanner). Slightly murkier are indirect costs, which are defined by default. These indirect costs Eur J Nucl Med Mol Imaging (2007) 34:625–627 DOI 10.1007/s00259-006-0308-y

Highlights of the Annual Congress of the European Association of Nuclear Medicine, Istanbul, 2005: the incremental value of nuclear medicine for patient management and care

The 2005 Annual Congress of the European Association of Nuclear Medicine (EANM) took place in Istanbul on October 15-19, under the chairmanship of Professor Hatice Durak. The programme was of excellent quality and represented a further step towards the achievement of a standardized EANM congress structure. A large industrial exhibition demonstrated the latest technological innovations and developments within the field. The congress was a great success: there were more than 4,000 participants, and 1,670 abstracts were received. Of these, 1,399 were accepted for oral or poster presentations, with a rejection rate of 16.2%. The original investigations presented were related to different areas of nuclear medicine, and addressed particularly advances in instrumentation and data processing, progress in radiochemistry and pharmacy, novel diagnostics and therapeutics, and new insights in well-established areas of clinical application, such as oncology, cardiology, neurology, psychiatry, endocrinology, paediatrics, nephrology, and infection and inflammation. It is noteworthy that a number of studies presented at this congress focussed on the quantitative interpretation of the imaging data and on pragmatic endpoints, such as adverse outcomes, and identified when nuclear medicine procedures achieved clinical effectiveness for patient care and management. These and many other studies presented at the congress demonstrate once more the crucial role that nuclear medicine has to play in contemporary medicine. This highlights lecture is only a brief summary of the large amount of data presented and discussed, which can be found in much greater detail in the congress proceedings book, published as volume 32, supplement 1 of the Eur J Nucl Med Mol Imaging in September 2005.