British Journal of Radiology 75 (2002),S39-S52 © 2002 The British Institute of Radiology
Where can FDG-PET contribute most to anatomical imaging problems?
J B Bingham
Radiological Sciences, 5th Floor TGH, Guy's Hospital, London SE1 9RT, UK
All the main cross-sectional anatomical imaging techniques, CT, MR and ultrasound, suffer from the same problem, that they lack specificity. Benign may be indistinguishable from malignant tumours and the appearances of infection frequently overlap those of malignant disease and vice-versa. A vast amount of data are presented in each image of a CT or MRI series and it is very easy to overlook significant involvement of nodes and organs by metastatic disease.
Nodal involvement by tumour is usually assessed using the size of the nodes. This is determined by criteria derived from data from the literature and is dependent on the anatomical site of the nodes. Small nodes are frequently involved by tumour and enlarged nodes may be affected by reactive hyperplasia. In the chest, 1 cm is usually regarded as the cut-off level [1]. Biopsy is usually required to confirm involvement of nodes by tumour but may be hazardous and unnecessary if there is involvement by tumour elsewhere in the body where it is more accessible to biopsy. Random biopsy may also be unrepresentative in large tumours where necrosis and variability of tumour histology may make fine needle biopsy and even core biopsy unrepresentative.
Following treatment of tumours, there may be considerable regression in size of the malignant lesion but a significant mass may remain, possibly comprising non-active tumour and a residual core of fibrosis. This is a particular problem with both lymphoma and tumours of testicular origin. In conditions where there are multiple lesions that have a malignant potential, it can be very difficult to recognize malignant transformation without the aid of a metabolic tracer. This is the situation with neurofibromatosis where, although the malignant potential of lesions is relatively low, the sheer number of tumours means that recognition of malignant transformation is extremely difficult, particularly since they have no distinguishing characteristics with conventional cross-sectional imaging (Figure 1
).
View larger version (103K):
[in this window]
[in a new window]
|
Figure 1. (a) Coronal STIR images in a patient with neurofibromatosis and pain in the left arm. A number of small neurofibromata are visible but no identifiable cause of pain. (b) An FDG-PET scan shows a high uptake tumour in the left brachial plexus. (c) A repeat MR scan with a body coil confirms the tumour. This was surgically excised and found to be a neurofibrosarcoma.
|
|
Whole body examination has become more practical with spiral CT, particularly with multislice imaging, and with improving technology MR has also become a potential whole body imaging technique. Nonetheless, FDG-PET imaging provides the ability to cover the whole body without increasing the radiation burden. However, PET scanning is not without its drawbacks. The spatial resolution is considerably inferior to all the other commonly available cross-sectional techniques and this is compounded by a lack of background information to provide anatomical localization. This is partly mitigated by incorporating transmission scans into the data but at a considerable cost in time. Combining a PET and CT scanner in the same machine goes a long way to solving this problem as has already been demonstrated [2]. The technique has the advantage of not only improving resolution but also increasing the speed of imaging by obviating the need for a transmission scan. However, the majority of problems can be solved by direct inspection of anatomical imaging in conjunction with the nuclear study. Unfortunately, these are frequently obtained at different times with non-matching orientation, and in the absence of patient archiving systems (PACS) the anatomical imaging data may not be available at the time of interpretation of the images.
The problems of anatomical imaging are well illustrated in carcinoma of the lung where mediastinal involvement of nodes is frequent, access for biopsy is limited and there is considerable sampling error. Nodules in the lungs are often present and are poorly characterized by CT. There may be early spread to the ipsilateral lymph nodes and distant metastasis is common, making lesions incurable by surgery at presentation. A number of papers attest to the high sensitivity and specificity of PET-FDG compared with CT [3–6]. It is not uncommon for a malignant lesion on one side to be accompanied by a nodule in the other lung, but this is not necessarily metastatic and PET scanning can confirm the likelihood of a benign lesion accounting for the nodule on the contralateral side. Ipsilateral lymph nodes may be involved but be normal in size or not appreciated on CT (Figure 2).
View larger version (65K):
[in this window]
[in a new window]
|
Figure 2. (a) CT scan through the chest of a patient with lung cancer showing a small right paratracheal node. (b) The FDG-PET scan shows high uptake in the primary tumour and (c) uptake in the right paratracheal node.
|
|
Similarly, adrenal lesions are common but are frequently due to benign adenomata, and PET scanning will show their probable aetiology. There may be early liver or bone involvement, which may be overlooked on conventional imaging (Figure 3). Mediastinal disease may be visible, but the patient is at high risk for biopsy and further information is required as to the likely nature of the lesion before biopsy is warranted (Figure 4).
View larger version (77K):
[in this window]
[in a new window]
|
Figure 3. (a). A whole body FDG-PET scan confirms a large tumour in the right lung. (b) Comparison of the PET scan and CT shows that there is also a lesion in the left chest wall and (c) in the liver which had not been appreciated when the CT scan was first read.
|
|
View larger version (54K):
[in this window]
[in a new window]
|
Figure 4. (a) This patient had had a previous lobectomy for carcinoma of the lung and now has a small anterior mediastinal nodule but was a high risk for biopsy. (b) The FDG-PET scan confirms high uptake in the nodule indicating that biopsy is worthwhile.
|
|
In general terms, benign lesions in the lung have 10–25% of the uptake of malignant lesions but there are causes of positive uptake including aspergillomas, tuberculosis and, predominantly in the United States, histoplasmosis. There tends to be a slight increase in false negative examinations in the lower lobes, presumably because of the increased motion of the lungs at the bases. Recent publications have emphasized the alteration in management, based on PET scanning, which can result in considerable cost savings to the health services [4].
The sensitivity of MR for the detection of tumours in the brain is very high but, nonetheless, specificity remains limited and there is considerable overlap between pathological entities. Tumour grading is difficult with both CT and MR and on occasion it may be difficult to be certain that a tumour is present. Both FDG-PET and methionine imaging can confirm the presence of tumour, direct a suitable site for biopsy and assess the grade. Figure 5 shows an example of a child subsequently shown to have neurofibromatosis who presented with multiple cranial nerve palsies. Signal changes in the MR scan are subtle and the neurosurgeon required further confirmation of tumour involvement before he was prepared to risk a biopsy in such a crucial area. The methionine scan clearly shows an avid tumour, tipping the balance in favour of biopsy, which confirmed the presence of a grade II fibrillary glioma.
View larger version (64K):
[in this window]
[in a new window]
|
Figure 5. (a) An axial T2W series through the brainstem of a 2 year old with neurofibromatosis and recent cranial nerve palsy shows swelling of the pons but with little signal change. (b) The sagittal T1W images through the midline confirmed the swelling but no enhancement. (c) A PET scan with methionine (above) and FDG (below) showed increased uptake, making biopsy likely to be of benefit despite the hazardous site. A fibrillary grade II astrocytoma was confirmed.
|
|
Contrast enhancement and mass effect on CT and MR are common following radiotherapy treatment of tumours and may be very difficult to distinguish from recurrent disease. Both tracers are utilized to confirm the presence of metabolically active tumour recurrence whilst radiation necrosis is negative [7, 8]. This is exemplified in Figure 6 where a 68-year-old man having previously presented with a glioblastoma multiforme had re-presented with deterioration following biopsy and radiation therapy. The appearances of the MR scan had not changed over a 4 month period but increased uptake of methionine in the rim of the lesion indicated that radiation fibrosis was unlikely and that recurrent tumour was the probable cause of his deterioration. He subsequently died from the effects of progressive tumour.
View larger version (54K):
[in this window]
[in a new window]
|
Figure 6. (a) This man had a previously radiotherapy treated glioblastoma multiforme in the left parieto-occipital region. A T1W axial series showed a ring-enhancing lesion at the site of the previous tumour. The concern was whether this was recurrence of the tumour or radiation necrosis. (b) The PET scan (methionine above and FDG below) showed uptake in the abnormality, indicating it was likely to be tumour recurrence. This was confirmed by biopsy and the patient subsequently died from progression of the tumour.
|
|
In HIV-related disease, the distinction on CT and MR between non-malignant disease and malignant complications such as lymphoma may be extremely difficult. Progressive multifocal leucoencephalopathy (PML) and infective causes of disease such as toxoplasmosis may mimic lymphoma and since the treatment is different and biopsy difficult, PET scanning, with its ability to identify malignant lymphoma, may be extremely helpful. Figure 7 shows a patient with progressive changes in the left cerebellar hemisphere extending into the brain stem from the left that was negative on PET scanning and proved to be PML on biopsy. It is important to view the CT or MR in conjunction with the PET scan, as displacement of intra-cerebral structures such as the basal ganglia by mass effect or oedema may contribute to misinterpretation of tracer uptake.
View larger version (90K):
[in this window]
[in a new window]
|
Figure 7. (a) An axial T2W image in an HIV positive male shows abnormal signal in the left cerebellar peduncle extending into the pons. This rapidly progressed. The coronal T1W image after contrast (b) did not show significant enhancement and the concern was whether this might be lymphoma rather than progressive multifocal leucoencephalopathy. (c) The FDG-PET scan shows no significant uptake, indicating that progressive multifocal leucoencephalopathy was more likely. This was confirmed by biopsy.
|
|
Evaluation of malignant melanoma by FDG-PET has proved to be very successful in the recognition of both primary lesions and of metastatic disease, as well as identification of recurrence at the operative site following excision, as there is very high uptake of FDG in melanine-containing tumours [9].
All malignant diseases of the gastrointestinal tract may be evaluated by PET scanning. Accurate staging of the primary lesion is impractical with tracers as the disease is organ confined and the resolution of PET is not adequate to determine the wall involvement. However, from the point of view of nodal metastatic disease and recurrence, FDG-PET has a large role to play and has been shown to be both more sensitive and more specific than both CT and endoscopic ultrasound for nodal disease [10, 11]. Similar results have been obtained for gastric [12] and colonic cancer (Figure 8). Secondary spread to the liver is detected with a higher sensitivity than either CT or conventional MR, even with intravenous contrast medium. Early recurrence can be detected particularly following successful laser treatment, as demonstrated in Figure 9, in a patient with a previous laser-induced defect, where it was difficult to determine whether there was new disease around the treated site. The presence of tracer uptake confirms that this was recurrence. In colorectal cancer, a PET scan is indicated prior to resection of an isolated recurrence to ensure that it is indeed isolated, and PET scanning can be very effective for detecting the site of recurrence in patients with rising serum markers for disease [13]. Post-operative changes following excision of rectal tumours can be extremely difficult to distinguish by both CT and MR from tumour recurrence, and again PET scanning plays a large role in this distinction.
View larger version (80K):
[in this window]
[in a new window]
|
Figure 8. (a) A PET scan was performed on a patient with previously resected gastric carcinoma because of subcarinal abnormality which was thought could be a post-operative collection and (b) a granuloma at the right hilum. Both abnormalities showed intense uptake of FDG, indicating metastatic disease.
|
|
View larger version (83K):
[in this window]
[in a new window]
|
Figure 9. (a) A contrast enhanced CT of the liver in a patient who had received laser therapy to a solitary right lobe metastasis showed a low density laser defect. (b) and (c) T1W images before and after intravenous Gd-DTPA show two faint rounded enhancing lesions in the left lobe. (d) A PET-FDG scan confirms multiple high uptake metastases.
|
|
PET scanning has been used in the evaluation of patients with disseminated malignant disease from an unknown primary, and in general terms it is possible to identify the primary in about a quarter of the patients where lesions are detected elsewhere [14]. The usual origins are breast, lung, colon and nasopharynx but unfortunately survival is little changed by this information. The same applies to patients with cervical node metastases where again the primary can be identified in about 25% and may help guide biopsy [15]. PET scanning is particularly useful in recurrent disease of the neck where the distinction between post-operative changes, treatment fibrosis and recurrent tumour can be extremely difficult. Figure 10 shows an example of a patient with a previously excised Hurtle cell carcinoma with a rising thyroglobulin level, and the PET scan shows very clearly the positive area, which was subsequently excised.
View larger version (51K):
[in this window]
[in a new window]
|
Figure 10. (a) A CT scan through the neck of a man with a previously resected Hurtle cell carcinoma was equivocal for recurrence in the right thyroid bed. (b) A PET-FDG scan confirms high intensity uptake in the right side with recurrence confirmed by surgery.
|
|
Image registration may be effective for the identification of anatomical landmarks. This can be achieved using skin markers and matching them. More recent computer programmes use anatomical structures for matching and these can be very accurate, although care has to be taken to ensure that the registration has occurred effectively (Figure 11). This paves the way for image guided biopsy of the most metabolically active part of tumours so that a representative sample can be obtained for histological analysis (Figures 12 and 13).
View larger version (47K):
[in this window]
[in a new window]
|
Figure 11. An FDG-PET registration study in a patient with recurrent tumour in the right maxilla showing good matching of isotope uptake with the tumour visible in the T1W MR images.
|
|
View larger version (96K):
[in this window]
[in a new window]
|
Figure 12. (a) and (b) Sagittal STIR and axial fat suppressed T1W images in a patient with a small mass in the left thigh. This shows only mild uptake of FDG in (c) and was confirmed surgically to be a small haemangioma.
|
|
View larger version (71K):
[in this window]
[in a new window]
|
Figure 13. (a) Axial fat-suppressed post-contrast T1W images after contrast show a large necrotic mass in the left thigh with an enhancing rim. (b) Local views of the tumour in the thigh confirm a highly metabolic margin. (c) Whole body views show that the patient has multiple deposits in the lungs.
|
|
In summary, FDG-PET can detect disease in patients where anatomical imaging has difficulties, including those with complicating disease, and is particularly valuable in pre-operative evaluation of patients prior to potentially disfiguring surgery for malignant disease where the presence of metastatic spread precludes surgical cure. The distinction between post-operative change and radiotherapy effects from recurrent tumour can be effectively made with PET scanning.
 |
References
|
|---|
- Libshitz HI, McKenna RJ. Mediastinal lymph node size in lung cancer. Am J Roentgenol 1984;143:715–8.[Abstract/Free Full Text]
- Townsend DW, Beyer T. A combined PET/CT scanner: the path to true image fusion. Br J Radiol 2002;75:S24–S30.[Abstract/Free Full Text]
- Gupta NC, Graeber GM, Rogers JS, Bishop HA. Comparative efficacy of positron emission tomography with FDG and computed tomographic scanning in preoperative staging of non-small cell lung cancer. Ann Surg 1999;229:286–91.[Medline]
- Saunders CAB, Dussek JE, O'Doherty MJ, Maisey MN. Evaluation of fluorine-18-fluorodeoxyglucose whole body positron emission tomography imaging in the staging of lung cancer. Ann Thorac Surg 1999;67:790–7.[Abstract/Free Full Text]
- Vansteenkiste JF, De Leyn PR, Vershakelen JA, Mortelmans LA. Mediastinal lymph node staging with FDG-PET scan in patients with potentially operable non-small cell lung cancer. Chest 1997;112:1480–6.[Abstract/Free Full Text]
- Weng E, Tran L, Rege S, Safa A, Sadeghi A, Juilliard G, et al. Accuracy and clinical impact of mediastinal lymph node staging with FDG-PET imaging in potentially resectable lung cancer. Am J Clin Oncol 200;23:47–52.[Medline]
- Kim EE, Chung SK, Haynie TP, et al. Differentiation of residual or recurrent tumours from post-treatment changes with F-18 FDG. Radiographics 1992;12:269–79.[Abstract]
- Di Chiro G, Oldfield E, Wright DC, et al. Cerebral necrosis after radiotherapy and/or intraarterial chemotherapy for brain tumours: PET and neuropathological studies. Am J Roentgenol 1988;150:189–97.[Abstract/Free Full Text]
- Holder WD, White RL, Zuger JH, Easton EJ, Greene FL. Effectiveness of positron emission tomography for the detection of melanoma metastases. Ann Surg 1998;227:769–71.
- Choi JY, Lee KH, Shim YM, Lee KS, Kim JJ, Kim SE, et al. Improved detection of individual nodal involvement in squamous cell carcinoma of the oesophagus by FDG-PET. J Nucl Med 2000;41:808–15.[Abstract/Free Full Text]
- Kobori O, Kirihara Y, Kosaka N, Hara T. Positron emission tomography of esophageal carcinoma using (11) C-choline and (18) F-fluorodeoxyglucose: a novel method of preoperative lymph node staging. Cancer 1999;86:1638–48.[Medline]
- Lerut T, Flamen P, Ectors N, Van Cutsem E, Peeters M, Hiele M, et al. Histopathologic validation of lymph node staging with FDG-PET scan in cancer of the esophagus and gastoesophageal junction: a prospective study based on primary surgery with extensive lymphadenectomy. Ann Surg 2000;232:743–52.[Medline]
- Imdahl A, Reinhardt MJ, Nitzsche EU, Mix M, Dingeldey A, Einert A, et al. Impact of 18F-FDG-positron emission tomography for decision making in colorectal cancer recurrences. Langenbecks Arch Surg 2000;385:129–34.[Medline]
- Kole AC, Nieweg OE, Pruim J, Hoekstra HJ, Koops HS, Roodenburg JL, et al. Detection of unknown occult primary tumors using positron emission tomography. Cancer 1998;82:1160–6.[Medline]
- Braams JW, Pruim J, Kole AC, Nikkels PG, Vaalburg W, Vermey A, et al. Detection of unknown primary head and neck tumors by positron emission tomography. Int J Oral Maxillofac Surg 1997;26:112–5.[Medline]
Top
References
