Volume 41, Issue 1 , Pages 5-8, January 2012
Navigation-assisted localisation and resection of subclinical metastatic malignant melanoma of unknown primary based on 18-fluorodeoxyglcose positron emission tomography computed tomography fusion imaging
Article Outline
Abstract
A technical application using a stereotactic navigation system with fusion images of [18F]-2-fluorodeoxyglucose (FDG) positron tomography and computed tomography (PET-CT) in the case of a metastatic melanoma of unknown primary site is described. A 50-year-old woman presented with a slow, growing level V neck lump which was cytologically proved to be a metastatic melanoma despite the absence of prior or existing history of skin malignancy. Whilst detailed physical examination failed to yield the site of the primary lesion, full body FDG-PET images isolated FDG-avid subclinical scalp lesions. Fused PET-CT data provided the navigation system with accurate localisation of the subclinical metastatic lesion. Histopathological examination of the navigation-guided resection specimen confirmed that the lesion was excised with acceptable margins. This case illustrates the feasibility of navigation-assisted resection of a subclinical malignant melanoma lesion and may have a role to play in the management of the melanoma of unknown primary.
Key words: navigation surgery, unknown primary, metastatic melanoma, PET-CT, head, neck
Positron emitted tomography (PET) with glucose analogue 18-fluorodeoxyglucose (FDG) is a non-invasive, metabolic imaging modality5. Tumour localisation is based on the avid FDG uptake by malignant cells, which have a much higher glucose utilisation rate than normal cells. It is often used to detect the unknown primary with occult cervical metastasis5, including malignant melanoma of unknown primary6. When FDG-PET images are used alone, it is extremely difficult to establish the exact three-dimensional (3D) position of the FDG avid area, limiting its value in disease staging and treatment5. In relation to malignant melanoma, fusion of FDG-PET and computed tomography (CT) images is more precise than conventional FDG-PET alone when used in diagnosis and staging7.
The goal of intraoperative localisation of a pre-determined area on imaging is now achievable with a degree of accuracy through navigation systems. The authors describe a technique utilising navigation-assisted surgery based on fused FDG PET-CT images to localise a subclinical metastatic melanoma lesion, detected only on imaging.
Case report
A 50-year-old woman presented with a slowly enlarging 3
cm lump in the left level 5 neck area. The resultant ultrasound guided, fine needle aspiration cytology indicated metastatic melanoma despite the absence of any current or previous malignant cutaneous lesions. A whole body FDG PET-CT was requested when systematic examination of the skin, oral, otolaryngological, urogenital system and eyes were found to be normal.
PET-CT data acquisition
The whole body PET-CT was performed using a hybrid PET-CT system (Biograph 64 TruePoint PET-CT, Siemens, Germany) with an 18-FDG radiotracer. The patient fasted 6h prior to the intravenous injection of 18-FDG titred to body weight and data acquisition commenced 1h post-injection. The CT scanning protocol used was: slice thickness 5mm; pitch 0.8mm; rotation 0.5s; and field of view 700mm. The PET scanner had a field of view that matched the CT scan and the total acquisition time for PET-CT head and neck data was 7.5min. The PET-CT image showed several FDG-avid lesions (Fig. 1). A scalp lesion was seen in the left parietal region and the other represented the positive node in the level 5 region of the left neck.
Fig. 1.
PET-CT images showing FDG-avid lesions in the temporal area (red arrows) and the level 5 neck (yellow arrows).
Navigation-guided surgery
The navigation system (VoNavix, IVS Technology GmbH, Chemnitz, Germany) used was an optical-based system, specialised for craniofacial stereotactic navigation. The hardware consisted of a 3D camera with an infrared light source, passive markers (located on the patient and the instrument pointer) and a computer workstation. The system software (Voxim, IVS Technology GmbH, Chemnitz, Germany) constructs a virtual 3D dataset using acquired CT or MR images. The patient's PET-CT images were exported as diacom files and image fusion was performed using the navigation software. Using segmentation, the area of enhanced radiotracer activity was isolated (Fig. 2), allowing its identification on the virtual 3D dataset model.
Fig. 2.
Using segmentation, the area of enhanced radiotracer activity was isolated (marked area behind the ear).
Following standard skin preparation and draping, three passive markers were fixed onto the contralateral skull. Patient-to-image registration was performed using three match points on fixed soft tissue landmarks (tragus, alar and medial canthus). The target area was identified purely by stereotactic navigation and this was marked out with surgical marking ink (Fig. 3). Resection of the marked area (diameter 18mm) was performed allowing a 1cm margin around the target area and this was sent for histopathological examination. The patient underwent left modified radical neck dissection and the postoperative recovery period was uneventful.
Fig. 3.
Using the instrument pointer, the target area was identified purely by stereotactic navigation and was marked out with surgical marking ink. Note the three passive markers fixed on the contralateral skull.
Technique validation
Histopathological examination of the scalp resection specimen (diameter 36mm; depth 11mm) was performed. Within the subcutaneous fat and muscle layers, at a depth of 6mm, a small lymph node (maximum diameter 12mm) was found and this was completely replaced by metastatic malignant melanoma. This lymph node was situated towards the anterior edge of the resected scalp specimen, approximately 7mm from the anterior margin, 4.1mm from the superior margin and 3mm from the deep margin (Fig. 4). On the basis of pathological findings, the authors conclude that this technique facilitated complete resection of a subclinical lesion.
Fig. 4.
Resection material. The specimen measured 36
mm across and was 11mm deep. At a depth of 6mm, in the subcutaneous and fat layers, a small lymph node with a maximum diameter of 12mm was found. It was completely replaced by metastatic malignant melanoma. This lymph node was situated towards the anterior aspect of the specimen, approximately 7mm from the anterior margin, 4mm from the superior margin and 3mm from the deep margin.
Discussion
Navigation-assisted oncology surgery based on PET-CT images has been described previously as facilitating biopsy2, detecting local recurrence in areas of disrupted anatomy from previous treatment3 and providing control of resection margins4. The majority involved the diagnosis of squamous cell carcinoma3, 4 apart from one case of aspergilloma2. The present report describes navigation-assisted oncology surgery using PET-CT images in the resection of a subclinical, metastatic malignant melanoma of unknown primary. Whilst the histological examination showed that the malignant focus was excised completely, the planned surgical margin around the lesion was less than the recommended surgical clearance margin. The reliability of this technique depends on the individual accuracy of the hybrid PET-CT system and the navigation system to provide accurate intraoperative localisation.
The accuracy of the navigation system is largely determined by registration accuracy. Errors introduced during registration are sometimes unavoidable, exemplified by feducial registration errors (FRE), where the marker position does not tally accurately with image or patient coordinates following patient-to-image registration, which occurs in all navigation systems1. The authors’ technique is based on a markerless pair-point method, where anatomical landmarks are identified on both the patient and the image data, which was preferred to permit greater operating freedom. As the planned surgical procedure involved mainly soft tissue work, relatively fixed soft tissue landmarks were chosen. Utilising superficial landmarks could introduce inaccuracies to the system, such as a feducial localisation error (FLE), as these points may be subject to structural deformation from oedema or displacement1. The authors attempted to circumvent this by choosing multiple registration points that are located well away from the target and by placing the fixed passive markers on the contralateral side to avoid distortion of soft tissue landmarks. Apart from the issues relating to registration accuracy of the navigation system, underestimation of the tumour shape and volume has been reported with different types of segmentation tools8. The authors are unable to comment whether this element was contributory to the relatively large shape of the target area delineated in the present case when this is compared with the histological findings.
The melanoma lesion identified by histopathological examination was located just outside the target area indicated by the PET-CT data. There is a difference of 12mm between the centre of the target area and the centre of the actual lesion. This inaccuracy may be the combined result of FLE, FRE and target registration error (recorded as 5mm on the navigation system). Whilst registration accuracy may be further improved, for example by increasing PET-CT image resolution, and the number and choice of registration points, this study indicates that an additional margin representing the potential error around the target area should be considered to allow sufficient oncological clearance. This technique may provide an alternative to ultrasound-guided surgery in occult melanoma9.
Conflict of interest
None declared.
Funding
None.
Ethical approval
None.
References
- . Image-to-patient registration techniques in head surgery. Int J Oral Maxillofac Surg. 2006;35:1081–1095
- . Case report: fusion of positron emission tomography (PET) and computed tomography (CT) images for image-guided endoscopic navigation in maxillofacial surgery: clinical application of a new technique. J Craniomaxillofac Surg. 2007;35:322–328
- . F-18 positron emission tomography and computed tomography image-fusion for image-guided detection of local recurrence in patients with head and neck cancer using a 3-dimensional navigation system: a preliminary report. J Oral Maxillofac Surg. 2008;66:193–200
- . Intraoperative control of resection margins in advanced head and neck cancer using a 3D-navigation system based on PET/CT image fusion. J Craniomaxillofac Surg. 2010;38:589–594
- . Recommendations on the use of 18F-FDG PET in oncology. J Nucl Med. 2008;49:480–508
- . Clinical applications of fluorodeoxyglucose-positron emission tomography in the management of malignant melanoma. Curr Opin Oncol. 2005;17:154–159
- . Multimethod imaging, staging, and spectrum of manifestations of metastatic melanoma. Clin Radiol. 2011;66:224–236
- . Comparison of five segmentation tools for 18F-fluoro-deoxy-glucose-positron emission tomography-based target volume definition in head and neck cancer. Int J Radiat Oncol Biol Phys. 2007;69:1282–1289
- . Detection of occult melanoma by preoperative positron emission tomography-computed tomography and intraoperative ultrasonography. J Cutan Med Surg. 2010;14:130–135
PII: S0901-5027(11)01445-7
doi:10.1016/j.ijom.2011.09.015
© 2011 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Inc. All rights reserved.
Volume 41, Issue 1 , Pages 5-8, January 2012
