الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 124 |  |  |
| | | from | 124 | | |
Abstract
Head and neck cancers are common and one of the leading causes of death worldwide. Tumor resection with an adequate safety margin is considered the most prognostic factor for the patients’ overall survival. Safety margin is achieved either by frozen section histopathology intraoperative or by using computer assisted surgery (CAS). CAS include using tumor resection guides based on the imaging modality used for patient assessment. Using PET/CT for fabrication of resection guide is more accurate than single imaging modality as CT and MRI. This is due to better delineation of the anatomy and metabolic tumor activity together and less interobserver variability in tumor segmentation.
Patients and methods: Eight patients were included in the study for evaluation of evaluate the accuracy of PET-CT resection guide in reducing the positive tumor margins after resection. Also to evaluate the accuracy of PET-CT resection guide in transferring the virtual planning of resection to the surgical field. For all the patients preoperative PET/CT were obtained and integrated into the software for virtual planning and guides fabrications. All the cases undergone computer assisted resection, then postoperative CT was obtained and rendered 3D model to be superimposed on the preoperative record for assessment of the virtual planning accuracy.
Results: One out of eight patients did not show any isotope uptake in PET/CT. Two out of the remaining seven patients, we could not fabricate a resection guide. The remainder five cases we used 4 guides. All of these guides showed 100% accuracy in obtaining the safety margin from the first resection attempt. The frozen sections showed perfect agreement (100 %) with the final pathology reports. The accuracy of transferring the digital planning to the actual surgery was very high, with a median accuracy of 99.3% and a median Delta of 0.1 mm. The Delta and accuracy were significantly different between the different planes (p-value: <0.001 and 0.019, respectively). The Frankfort horizontal plane (FHP) had the smallest median Delta (0 mm) and the highest median accuracy (99.8%). Even the least accurate plane, the coronal plane (CP), showed high median accuracy (98.9%) and arguably a clinically insignificant Delta (0.8 mm).
Conclusion: PET-CT based resection guides confers high accuracy and precision for maxillofacial malignant tumors. It significantly decreased the positive or close tumor margins. The currently used technique offered an accurate workflow in transferring the virtual surgical planning of resection to the surgical field.